you com essay

You.com launches an AI-powered writing tool powered by OpenAI

• Share on Facebook
• Share on LinkedIn

Join us in Atlanta on April 10th and explore the landscape of security workforce. We will explore the vision, benefits, and use cases of AI for security teams. Request an invite here.

Just a few months ago, Bryan McCann and Richard Socher, the former chief scientist at Salesforce, launched You.com , a search engine that leverages AI to understand search queries, rank the results, and parse the queries into different languages (including programming languages). The platform summarizes information from across the web and is extensible with built-in search apps, like apps for Yelp and Twitter, so that users can complete tasks without having to leave the results page.

In its quest to recalibrate expectations around search engines, You.com is today launching a search app built in collaboration with OpenAI that generates snippets — or even documents — of text when given a prompt. Socher calls it a “personal AI writer.”

“[T]his is our first foray into what we call the app store, which doesn’t optimize for you spending as much time on there so we can sell you advertisement, but for you, actually getting stuff done,” Socher told VentureBeat in a phone interview. “[It’s perfect for] if you have writer’s block.”

You.com’s new tool is powered by the same technology behind OpenAI’s GPT-3 , an AI language system that can generate human-like poetry, emails, recipes, short stories, movie scripts, and more. Socher wasn’t keen to disclose many of the technical details, but described You.com’s relationship with OpenAI as a “partnership” and the model underpinning the tool as “very similar” to GPT-3. (When contacted for comment, an OpenAI spokesperson said that YouWrite is powered by GPT-3 — specifically the recently-released InstructGPT models — through its API.)

The AI Impact Tour – Atlanta

To use You.com’s writing assistant, called YouWrite, users type a query like “How to write an essay” into the search engine’s search bar and click the magnifying glass icon. Up pops a widget with options that let the user specify the length (e.g., paragraph), the audience or receiver (e.g., students, teachers, or marketers), tone (e.g., persuasive), and the content of the message (e.g., “three paragraphs on the Civil War”) they want YouWrite to generate.

“We want to, basically, create this AI-powered writing system to help people be more productive, but also being controlled — you can decide what it should write,” Socher said. “We want to put people into control of the AI to make them more efficient.”

In a demo, Socher showed how YouWrite can be prompted to write paragraphs explaining “why dogs are awesome,” a blog post about a new search engine, or a boilerplate rejection letter for a job candidate (complete with a placeholder for the candidate’s name). While VentureBeat wasn’t given an opportunity to test the tool itself — Socher entered the prompts during a Zoom call — the quality of the text seemed at least on par with output from GPT-3 and other sophisticated language systems.

Of course, with any AI-powered language system, there’s a risk that the system might become susceptible to bias and toxicity. Language systems such as GPT-3 learn to “write” by analyzing huge chunks of text from websites, including from problematic sources advancing conspiracy theories, misinformation, racism, sexism, ageism, and ableism. OpenAI itself notes that biased datasets can lead to placing words like “naughty” or “sucked” near female pronouns and “Islam” near words like “terrorism.”

Socher claims that YouWrite prevents problematic outputs using filters and other techniques, like human feedback , on the backend. We’ll have to see how well the system holds up once it’s made public, but during the demo, typing the prompt “why jews are bad” yielded the message “We’re sorry, but we can’t return a good completion for your request.” YouWrite also seemed to be able to detect when its output might contain sensitive content, such as references to violence, and append a warning label.

You.com will offer YouWrite for free to start, but frequent users and those who use it to generate longer outputs (think essays) will eventually have to pay for the privilege. Socher says that pricing hasn’t been decide yet, but will be “a lot cheaper” than other AI-powered writing tools on the market, like Jasper and CopyAI.

“I think it’s really important for search engines like ours to be the best place to kind of explore this kind of new technology — to move away from, ‘Here’s a list of links that’s getting cluttered and full of ads,'” Socher said. “I think that ultimately, if you want to be a writer and have a search engine that helps you do research, summarize the web, and also get something on the page, You.com is going to be your best search engine.”

Stay in the know! Get the latest news in your inbox daily

By subscribing, you agree to VentureBeat's Terms of Service.

Thanks for subscribing. Check out more VB newsletters here .

An error occured.

You.Com: A New Frontier in AI-Infused Search Engines

This post describes a new search engine called You.com that leverages AI to understand search queries, rank results, and parse queries into different languages, including programming languages. While it still lacks the cohesion of larger competitors, this platform is a promising demonstration of what the future of AI could look like.

You.com is a search engine launched by Bryan McCann and Richard Socher that leverages AI to understand search queries, rank results, and parse queries into different languages, including programming languages. The platform uses similar technology to its partner, OpenAI, to understand users’ queries, identify patterns, and predict results that satisfy the user’s search intent.

Here are some of the important tools that You.com offers:

YouChat: With similar functionality to ChatGPT, it generates text responses resembling human responses to human prompts on any topic. YouChat can learn and summarize from websites and become an expert that a user can ask follow-up questions.

YouWrite: This tool has a writing feature with a mini-interface similar to popular writing apps like Writesonic. It can be used to write essays and generate tailored content for emails and social media posts.

YouCode: Aimed at developers, YouCode surfaces the best results for a code query from the user. It is a user-friendly integration of Google search and StackOverflow.

CodeComplete: With similar functionality to GitHub Copilot, it offers a series of pre-filled code samples for your query. It can generate decent code in Python, JavaScript, C++, and other languages.

YouImagine: Learning from your plain language description, it can produce realistic visuals and artwork. It would be nicer if it could remove any undesired elements from the generated images to make them look exactly as the user wants.

AI Image Resolution Enhancer: With the support of AI models like Stable diffusion 1.5 and 2.0, Open Journey, and Anime, it can generate 4x image resolution without losing image quality.

You.com is a conversational search engine that explores the web and detailed website contents, along with generating complementary content in the forms of art, code, and text within seconds. However, it still feels like a collection of parts without the absolutely airtight cohesion of its larger competitors in the market. But it summarizes information across the web with an integrated chatbot and is extensible with built-in search apps like Yelp and Twitter, so that users can complete tasks without leaving the results page. While it’s too early to make a revolution in the search engine business, You.com is a promising and interesting demonstration of what the future of AI could look like.

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base
• The four main types of essay | Quick guide with examples

The Four Main Types of Essay | Quick Guide with Examples

Published on September 4, 2020 by Jack Caulfield . Revised on July 23, 2023.

An essay is a focused piece of writing designed to inform or persuade. There are many different types of essay, but they are often defined in four categories: argumentative, expository, narrative, and descriptive essays.

Argumentative and expository essays are focused on conveying information and making clear points, while narrative and descriptive essays are about exercising creativity and writing in an interesting way. At university level, argumentative essays are the most common type. 

In high school and college, you will also often have to write textual analysis essays, which test your skills in close reading and interpretation.

Instantly correct all language mistakes in your text

Upload your document to correct all your mistakes in minutes

Table of contents

Argumentative essays, expository essays, narrative essays, descriptive essays, textual analysis essays, other interesting articles, frequently asked questions about types of essays.

An argumentative essay presents an extended, evidence-based argument. It requires a strong thesis statement —a clearly defined stance on your topic. Your aim is to convince the reader of your thesis using evidence (such as quotations ) and analysis.

Argumentative essays test your ability to research and present your own position on a topic. This is the most common type of essay at college level—most papers you write will involve some kind of argumentation.

The essay is divided into an introduction, body, and conclusion:

• The introduction provides your topic and thesis statement
• The body presents your evidence and arguments
• The conclusion summarizes your argument and emphasizes its importance

The example below is a paragraph from the body of an argumentative essay about the effects of the internet on education. Mouse over it to learn more.

A common frustration for teachers is students’ use of Wikipedia as a source in their writing. Its prevalence among students is not exaggerated; a survey found that the vast majority of the students surveyed used Wikipedia (Head & Eisenberg, 2010). An article in The Guardian stresses a common objection to its use: “a reliance on Wikipedia can discourage students from engaging with genuine academic writing” (Coomer, 2013). Teachers are clearly not mistaken in viewing Wikipedia usage as ubiquitous among their students; but the claim that it discourages engagement with academic sources requires further investigation. This point is treated as self-evident by many teachers, but Wikipedia itself explicitly encourages students to look into other sources. Its articles often provide references to academic publications and include warning notes where citations are missing; the site’s own guidelines for research make clear that it should be used as a starting point, emphasizing that users should always “read the references and check whether they really do support what the article says” (“Wikipedia:Researching with Wikipedia,” 2020). Indeed, for many students, Wikipedia is their first encounter with the concepts of citation and referencing. The use of Wikipedia therefore has a positive side that merits deeper consideration than it often receives.

Prevent plagiarism. Run a free check.

An expository essay provides a clear, focused explanation of a topic. It doesn’t require an original argument, just a balanced and well-organized view of the topic.

Expository essays test your familiarity with a topic and your ability to organize and convey information. They are commonly assigned at high school or in exam questions at college level.

The introduction of an expository essay states your topic and provides some general background, the body presents the details, and the conclusion summarizes the information presented.

A typical body paragraph from an expository essay about the invention of the printing press is shown below. Mouse over it to learn more.

The invention of the printing press in 1440 changed this situation dramatically. Johannes Gutenberg, who had worked as a goldsmith, used his knowledge of metals in the design of the press. He made his type from an alloy of lead, tin, and antimony, whose durability allowed for the reliable production of high-quality books. This new technology allowed texts to be reproduced and disseminated on a much larger scale than was previously possible. The Gutenberg Bible appeared in the 1450s, and a large number of printing presses sprang up across the continent in the following decades. Gutenberg’s invention rapidly transformed cultural production in Europe; among other things, it would lead to the Protestant Reformation.

A narrative essay is one that tells a story. This is usually a story about a personal experience you had, but it may also be an imaginative exploration of something you have not experienced.

Narrative essays test your ability to build up a narrative in an engaging, well-structured way. They are much more personal and creative than other kinds of academic writing . Writing a personal statement for an application requires the same skills as a narrative essay.

A narrative essay isn’t strictly divided into introduction, body, and conclusion, but it should still begin by setting up the narrative and finish by expressing the point of the story—what you learned from your experience, or why it made an impression on you.

Mouse over the example below, a short narrative essay responding to the prompt “Write about an experience where you learned something about yourself,” to explore its structure.

Since elementary school, I have always favored subjects like science and math over the humanities. My instinct was always to think of these subjects as more solid and serious than classes like English. If there was no right answer, I thought, why bother? But recently I had an experience that taught me my academic interests are more flexible than I had thought: I took my first philosophy class.

Before I entered the classroom, I was skeptical. I waited outside with the other students and wondered what exactly philosophy would involve—I really had no idea. I imagined something pretty abstract: long, stilted conversations pondering the meaning of life. But what I got was something quite different.

A young man in jeans, Mr. Jones—“but you can call me Rob”—was far from the white-haired, buttoned-up old man I had half-expected. And rather than pulling us into pedantic arguments about obscure philosophical points, Rob engaged us on our level. To talk free will, we looked at our own choices. To talk ethics, we looked at dilemmas we had faced ourselves. By the end of class, I’d discovered that questions with no right answer can turn out to be the most interesting ones.

The experience has taught me to look at things a little more “philosophically”—and not just because it was a philosophy class! I learned that if I let go of my preconceptions, I can actually get a lot out of subjects I was previously dismissive of. The class taught me—in more ways than one—to look at things with an open mind.

A descriptive essay provides a detailed sensory description of something. Like narrative essays, they allow you to be more creative than most academic writing, but they are more tightly focused than narrative essays. You might describe a specific place or object, rather than telling a whole story.

Descriptive essays test your ability to use language creatively, making striking word choices to convey a memorable picture of what you’re describing.

A descriptive essay can be quite loosely structured, though it should usually begin by introducing the object of your description and end by drawing an overall picture of it. The important thing is to use careful word choices and figurative language to create an original description of your object.

Mouse over the example below, a response to the prompt “Describe a place you love to spend time in,” to learn more about descriptive essays.

On Sunday afternoons I like to spend my time in the garden behind my house. The garden is narrow but long, a corridor of green extending from the back of the house, and I sit on a lawn chair at the far end to read and relax. I am in my small peaceful paradise: the shade of the tree, the feel of the grass on my feet, the gentle activity of the fish in the pond beside me.

My cat crosses the garden nimbly and leaps onto the fence to survey it from above. From his perch he can watch over his little kingdom and keep an eye on the neighbours. He does this until the barking of next door’s dog scares him from his post and he bolts for the cat flap to govern from the safety of the kitchen.

With that, I am left alone with the fish, whose whole world is the pond by my feet. The fish explore the pond every day as if for the first time, prodding and inspecting every stone. I sometimes feel the same about sitting here in the garden; I know the place better than anyone, but whenever I return I still feel compelled to pay attention to all its details and novelties—a new bird perched in the tree, the growth of the grass, and the movement of the insects it shelters…

Sitting out in the garden, I feel serene. I feel at home. And yet I always feel there is more to discover. The bounds of my garden may be small, but there is a whole world contained within it, and it is one I will never get tired of inhabiting.

Receive feedback on language, structure, and formatting

Professional editors proofread and edit your paper by focusing on:

• Academic style
• Vague sentences
• Style consistency

See an example

Though every essay type tests your writing skills, some essays also test your ability to read carefully and critically. In a textual analysis essay, you don’t just present information on a topic, but closely analyze a text to explain how it achieves certain effects.

Rhetorical analysis

A rhetorical analysis looks at a persuasive text (e.g. a speech, an essay, a political cartoon) in terms of the rhetorical devices it uses, and evaluates their effectiveness.

The goal is not to state whether you agree with the author’s argument but to look at how they have constructed it.

The introduction of a rhetorical analysis presents the text, some background information, and your thesis statement; the body comprises the analysis itself; and the conclusion wraps up your analysis of the text, emphasizing its relevance to broader concerns.

The example below is from a rhetorical analysis of Martin Luther King Jr.’s “I Have a Dream” speech . Mouse over it to learn more.

King’s speech is infused with prophetic language throughout. Even before the famous “dream” part of the speech, King’s language consistently strikes a prophetic tone. He refers to the Lincoln Memorial as a “hallowed spot” and speaks of rising “from the dark and desolate valley of segregation” to “make justice a reality for all of God’s children.” The assumption of this prophetic voice constitutes the text’s strongest ethical appeal; after linking himself with political figures like Lincoln and the Founding Fathers, King’s ethos adopts a distinctly religious tone, recalling Biblical prophets and preachers of change from across history. This adds significant force to his words; standing before an audience of hundreds of thousands, he states not just what the future should be, but what it will be: “The whirlwinds of revolt will continue to shake the foundations of our nation until the bright day of justice emerges.” This warning is almost apocalyptic in tone, though it concludes with the positive image of the “bright day of justice.” The power of King’s rhetoric thus stems not only from the pathos of his vision of a brighter future, but from the ethos of the prophetic voice he adopts in expressing this vision.

Literary analysis

A literary analysis essay presents a close reading of a work of literature—e.g. a poem or novel—to explore the choices made by the author and how they help to convey the text’s theme. It is not simply a book report or a review, but an in-depth interpretation of the text.

Literary analysis looks at things like setting, characters, themes, and figurative language. The goal is to closely analyze what the author conveys and how.

The introduction of a literary analysis essay presents the text and background, and provides your thesis statement; the body consists of close readings of the text with quotations and analysis in support of your argument; and the conclusion emphasizes what your approach tells us about the text.

Mouse over the example below, the introduction to a literary analysis essay on Frankenstein , to learn more.

Mary Shelley’s Frankenstein is often read as a crude cautionary tale about the dangers of scientific advancement unrestrained by ethical considerations. In this reading, protagonist Victor Frankenstein is a stable representation of the callous ambition of modern science throughout the novel. This essay, however, argues that far from providing a stable image of the character, Shelley uses shifting narrative perspectives to portray Frankenstein in an increasingly negative light as the novel goes on. While he initially appears to be a naive but sympathetic idealist, after the creature’s narrative Frankenstein begins to resemble—even in his own telling—the thoughtlessly cruel figure the creature represents him as. This essay begins by exploring the positive portrayal of Frankenstein in the first volume, then moves on to the creature’s perception of him, and finally discusses the third volume’s narrative shift toward viewing Frankenstein as the creature views him.

If you want to know more about AI tools , college essays , or fallacies make sure to check out some of our other articles with explanations and examples or go directly to our tools!

• Ad hominem fallacy
• Post hoc fallacy
• Appeal to authority fallacy
• False cause fallacy
• Sunk cost fallacy

College essays

• Choosing Essay Topic
• Write a College Essay
• Write a Diversity Essay
• College Essay Format & Structure
• Comparing and Contrasting in an Essay

 (AI) Tools

• Grammar Checker
• Paraphrasing Tool
• Text Summarizer
• AI Detector
• Plagiarism Checker
• Citation Generator

At high school and in composition classes at university, you’ll often be told to write a specific type of essay , but you might also just be given prompts.

Look for keywords in these prompts that suggest a certain approach: The word “explain” suggests you should write an expository essay , while the word “describe” implies a descriptive essay . An argumentative essay might be prompted with the word “assess” or “argue.”

The vast majority of essays written at university are some sort of argumentative essay . Almost all academic writing involves building up an argument, though other types of essay might be assigned in composition classes.

Essays can present arguments about all kinds of different topics. For example:

• In a literary analysis essay, you might make an argument for a specific interpretation of a text
• In a history essay, you might present an argument for the importance of a particular event
• In a politics essay, you might argue for the validity of a certain political theory

An argumentative essay tends to be a longer essay involving independent research, and aims to make an original argument about a topic. Its thesis statement makes a contentious claim that must be supported in an objective, evidence-based way.

An expository essay also aims to be objective, but it doesn’t have to make an original argument. Rather, it aims to explain something (e.g., a process or idea) in a clear, concise way. Expository essays are often shorter assignments and rely less on research.

The key difference is that a narrative essay is designed to tell a complete story, while a descriptive essay is meant to convey an intense description of a particular place, object, or concept.

Narrative and descriptive essays both allow you to write more personally and creatively than other kinds of essays , and similar writing skills can apply to both.

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

Caulfield, J. (2023, July 23). The Four Main Types of Essay | Quick Guide with Examples. Scribbr. Retrieved March 26, 2024, from https://www.scribbr.com/academic-essay/essay-types/

Is this article helpful?

Jack Caulfield

Other students also liked, how to write an argumentative essay | examples & tips, how to write an expository essay, how to write an essay outline | guidelines & examples, "i thought ai proofreading was useless but..".

I've been using Scribbr for years now and I know it's a service that won't disappoint. It does a good job spotting mistakes”

Tips for Writing an Effective Application Essay

How to Write an Effective Essay

Writing an essay for college admission gives you a chance to use your authentic voice and show your personality. It's an excellent opportunity to personalize your application beyond your academic credentials, and a well-written essay can have a positive influence come decision time.

Want to know how to draft an essay for your college application ? Here are some tips to keep in mind when writing.

Tips for Essay Writing

A typical college application essay, also known as a personal statement, is 400-600 words. Although that may seem short, writing about yourself can be challenging. It's not something you want to rush or put off at the last moment. Think of it as a critical piece of the application process. Follow these tips to write an impactful essay that can work in your favor.

1. Start Early.

Few people write well under pressure. Try to complete your first draft a few weeks before you have to turn it in. Many advisers recommend starting as early as the summer before your senior year in high school. That way, you have ample time to think about the prompt and craft the best personal statement possible.

You don't have to work on your essay every day, but you'll want to give yourself time to revise and edit. You may discover that you want to change your topic or think of a better way to frame it. Either way, the sooner you start, the better.

2. Understand the Prompt and Instructions.

Before you begin the writing process, take time to understand what the college wants from you. The worst thing you can do is skim through the instructions and submit a piece that doesn't even fit the bare minimum requirements or address the essay topic. Look at the prompt, consider the required word count, and note any unique details each school wants.

3. Create a Strong Opener.

Students seeking help for their application essays often have trouble getting things started. It's a challenging writing process. Finding the right words to start can be the hardest part.

Spending more time working on your opener is always a good idea. The opening sentence sets the stage for the rest of your piece. The introductory paragraph is what piques the interest of the reader, and it can immediately set your essay apart from the others.

4. Stay on Topic.

One of the most important things to remember is to keep to the essay topic. If you're applying to 10 or more colleges, it's easy to veer off course with so many application essays.

A common mistake many students make is trying to fit previously written essays into the mold of another college's requirements. This seems like a time-saving way to avoid writing new pieces entirely, but it often backfires. The result is usually a final piece that's generic, unfocused, or confusing. Always write a new essay for every application, no matter how long it takes.

5. Think About Your Response.

Don't try to guess what the admissions officials want to read. Your essay will be easier to write─and more exciting to read─if you’re genuinely enthusiastic about your subject. Here’s an example: If all your friends are writing application essays about covid-19, it may be a good idea to avoid that topic, unless during the pandemic you had a vivid, life-changing experience you're burning to share. Whatever topic you choose, avoid canned responses. Be creative.

6. Focus on You.

Essay prompts typically give you plenty of latitude, but panel members expect you to focus on a subject that is personal (although not overly intimate) and particular to you. Admissions counselors say the best essays help them learn something about the candidate that they would never know from reading the rest of the application.

7. Stay True to Your Voice.

Use your usual vocabulary. Avoid fancy language you wouldn't use in real life. Imagine yourself reading this essay aloud to a classroom full of people who have never met you. Keep a confident tone. Be wary of words and phrases that undercut that tone.

8. Be Specific and Factual.

Capitalize on real-life experiences. Your essay may give you the time and space to explain why a particular achievement meant so much to you. But resist the urge to exaggerate and embellish. Admissions counselors read thousands of essays each year. They can easily spot a fake.

9. Edit and Proofread.

When you finish the final draft, run it through the spell checker on your computer. Then don’t read your essay for a few days. You'll be more apt to spot typos and awkward grammar when you reread it. After that, ask a teacher, parent, or college student (preferably an English or communications major) to give it a quick read. While you're at it, double-check your word count.

Writing essays for college admission can be daunting, but it doesn't have to be. A well-crafted essay could be the deciding factor─in your favor. Keep these tips in mind, and you'll have no problem creating memorable pieces for every application.

What is the format of a college application essay?

Generally, essays for college admission follow a simple format that includes an opening paragraph, a lengthier body section, and a closing paragraph. You don't need to include a title, which will only take up extra space. Keep in mind that the exact format can vary from one college application to the next. Read the instructions and prompt for more guidance.

Most online applications will include a text box for your essay. If you're attaching it as a document, however, be sure to use a standard, 12-point font and use 1.5-spaced or double-spaced lines, unless the application specifies different font and spacing.

How do you start an essay?

The goal here is to use an attention grabber. Think of it as a way to reel the reader in and interest an admissions officer in what you have to say. There's no trick on how to start a college application essay. The best way you can approach this task is to flex your creative muscles and think outside the box.

You can start with openers such as relevant quotes, exciting anecdotes, or questions. Either way, the first sentence should be unique and intrigue the reader.

What should an essay include?

Every application essay you write should include details about yourself and past experiences. It's another opportunity to make yourself look like a fantastic applicant. Leverage your experiences. Tell a riveting story that fulfills the prompt.

What shouldn’t be included in an essay?

When writing a college application essay, it's usually best to avoid overly personal details and controversial topics. Although these topics might make for an intriguing essay, they can be tricky to express well. If you’re unsure if a topic is appropriate for your essay, check with your school counselor. An essay for college admission shouldn't include a list of achievements or academic accolades either. Your essay isn’t meant to be a rehashing of information the admissions panel can find elsewhere in your application.

How can you make your essay personal and interesting?

The best way to make your essay interesting is to write about something genuinely important to you. That could be an experience that changed your life or a valuable lesson that had an enormous impact on you. Whatever the case, speak from the heart, and be honest.

Is it OK to discuss mental health in an essay?

Mental health struggles can create challenges you must overcome during your education and could be an opportunity for you to show how you’ve handled challenges and overcome obstacles. If you’re considering writing your essay for college admission on this topic, consider talking to your school counselor or with an English teacher on how to frame the essay.

Related Articles

UPDATE: DC reaches deal to keep Capitals, Wizards at Capital One Arena until 2050

BREAKING NEWS: 2 bodies recovered from site of Baltimore bridge collapse

NEWS ALERT: Joe Lieberman, former Connecticut senator and Democratic VP candidate dies at 82

Should college essays touch on race? Some feel the affirmative action ruling leaves them no choice

The Associated Press

March 27, 2024, 2:51 PM

• Share This:
• share on facebook
• share on threads
• share on linkedin
• share on email

CHICAGO (AP) — When she started writing her college essay, Hillary Amofa told the story she thought admissions offices wanted to hear. About being the daughter of immigrants from Ghana and growing up in a small apartment in Chicago. About hardship and struggle.

Then she deleted it all.

“I would just find myself kind of trauma-dumping,” said the 18-year-old senior at Lincoln Park High School in Chicago. “And I’m just like, this doesn’t really say anything about me as a person.”

When the Supreme Court ended affirmative action in higher education, it left the college essay as one of few places where race can play a role in admissions decisions. For many students of color, instantly more was riding on the already high-stakes writing assignment. Some say they felt pressure to exploit their hardships as they competed for a spot on campus.

Amofa was just starting to think about her essay when the court issued its decision, and it left her with a wave of questions. Could she still write about her race? Could she be penalized for it? She wanted to tell colleges about her heritage but she didn’t want to be defined by it.

In English class, Amofa and her classmates read sample essays that all seemed to focus on some trauma or hardship. It left her with the impression she had to write about her life’s hardest moments to show how far she’d come. But she and some of her classmates wondered if their lives had been hard enough to catch the attention of admissions offices.

“For a lot of students, there’s a feeling of, like, having to go through something so horrible to feel worthy of going to school, which is kind of sad,” said Amofa, the daughter of a hospital technician and an Uber driver.

This year’s senior class is the first in decades to navigate college admissions without affirmative action . The Supreme Court upheld the practice in decisions going back to the 1970s, but this court’s conservative supermajority found it is unconstitutional for colleges to give students extra weight because of their race alone.

Still, the decision left room for race to play an indirect role: Chief Justice John Roberts wrote universities can still consider how an applicant’s life was shaped by their race, “so long as that discussion is concretely tied to a quality of character or unique ability.”

“A benefit to a student who overcame racial discrimination, for example, must be tied to that student’s courage and determination,” he wrote.

Scores of colleges responded with new essay prompts asking about students’ backgrounds. Brown University asked applicants how “an aspect of your growing up has inspired or challenged you.” Rice University asked students how their perspectives were shaped by their “background, experiences, upbringing, and/or racial identity.”

WONDERING IF SCHOOLS ‘EXPECT A SOB STORY’

When Darrian Merritt started writing his essay, he knew the stakes were higher than ever because of the court’s decision. His first instinct was to write about events that led to him going to live with his grandmother as a child.

Those were painful memories, but he thought they might play well at schools like Yale, Stanford and Vanderbilt.

“I feel like the admissions committee might expect a sob story or a tragic story,” said Merritt, a senior in Cleveland. “And if you don’t provide that, then maybe they’re not going to feel like you went through enough to deserve having a spot at the university. I wrestled with that a lot.”

He wrote drafts focusing on his childhood, but it never amounted to more than a collection of memories. Eventually he abandoned the idea and aimed for an essay that would stand out for its positivity.

Merritt wrote about a summer camp where he started to feel more comfortable in his own skin. He described embracing his personality and defying his tendency to please others. The essay had humor — it centered on a water gun fight where he had victory in sight but, in a comedic twist, slipped and fell. But the essay also reflects on his feelings of not being “Black enough” and getting made fun of for listening to “white people music.”

“I was like, ‘OK, I’m going to write this for me, and we’re just going to see how it goes,’” he said. “It just felt real, and it felt like an honest story.”

The essay describes a breakthrough as he learned “to take ownership of myself and my future by sharing my true personality with the people I encounter. … I realized that the first chapter of my own story had just been written.”

A RULING PROMPTS PIVOTS ON ESSAY TOPICS

Like many students, Max Decker of Portland, Oregon, had drafted a college essay on one topic, only to change direction after the Supreme Court ruling in June.

Decker initially wrote about his love for video games. In a childhood surrounded by constant change, navigating his parents’ divorce, the games he took from place to place on his Nintendo DS were a source of comfort.

But the essay he submitted to colleges focused on the community he found through Word is Bond, a leadership group for young Black men in Portland.

As the only biracial, Jewish kid with divorced parents in a predominantly white, Christian community, Decker wrote he constantly felt like the odd one out. On a trip with Word is Bond to Capitol Hill, he and friends who looked just like him shook hands with lawmakers. The experience, he wrote, changed how he saw himself.

“It’s because I’m different that I provide something precious to the world, not the other way around,” he wrote.

As a first-generation college student, Decker thought about the subtle ways his peers seemed to know more about navigating the admissions process . They made sure to get into advanced classes at the start of high school, and they knew how to secure glowing letters of recommendation.

If writing about race would give him a slight edge and show admissions officers a fuller picture of his achievements, he wanted to take that small advantage.

His first memory about race, Decker said, was when he went to get a haircut in elementary school and the barber made rude comments about his curly hair. Until recently, the insecurity that moment created led him to keep his hair buzzed short.

Through Word is Bond, Decker said he found a space to explore his identity as a Black man. It was one of the first times he was surrounded by Black peers and saw Black role models. It filled him with a sense of pride in his identity. No more buzzcut.

The pressure to write about race involved a tradeoff with other important things in his life, Decker said. That included his passion for journalism, like the piece he wrote on efforts to revive a once-thriving Black neighborhood in Portland. In the end, he squeezed in 100 characters about his journalism under the application’s activities section.

“My final essay, it felt true to myself. But the difference between that and my other essay was the fact that it wasn’t the truth that I necessarily wanted to share,” said Decker, whose top college choice is Tulane, in New Orleans, because of the region’s diversity. “It felt like I just had to limit the truth I was sharing to what I feel like the world is expecting of me.”

SPELLING OUT THE IMPACT OF RACE

Before the Supreme Court ruling, it seemed a given to Imani Laird that colleges would consider the ways that race had touched her life. But now, she felt like she had to spell it out.

As she started her essay, she reflected on how she had faced bias or felt overlooked as a Black student in predominantly white spaces.

There was the year in math class when the teacher kept calling her by the name of another Black student. There were the comments that she’d have an easier time getting into college because she was Black .

“I didn’t have it easier because of my race,” said Laird, a senior at Newton South High School in the Boston suburbs who was accepted at Wellesley and Howard University, and is waiting to hear from several Ivy League colleges. “I had stuff I had to overcome.”

In her final essays, she wrote about her grandfather, who served in the military but was denied access to GI Bill benefits because of his race.

She described how discrimination fueled her ambition to excel and pursue a career in public policy.

“So, I never settled for mediocrity,” she wrote. “Regardless of the subject, my goal in class was not just to participate but to excel. Beyond academics, I wanted to excel while remembering what started this motivation in the first place.”

WILL SCHOOLS LOSE RACIAL DIVERSITY?

Amofa used to think affirmative action was only a factor at schools like Harvard and Yale. After the court’s ruling, she was surprised to find that race was taken into account even at some public universities she was applying to.

Now, without affirmative action, she wondered if mostly white schools will become even whiter.

It’s been on her mind as she chooses between Indiana University and the University of Dayton, both of which have relatively few Black students. When she was one of the only Black students in her grade school, she could fall back on her family and Ghanaian friends at church. At college, she worries about loneliness.

“That’s what I’m nervous about,” she said. “Going and just feeling so isolated, even though I’m constantly around people.”

The first drafts of her essay focused on growing up in a low-income family, sharing a bedroom with her brother and grandmother. But it didn’t tell colleges about who she is now, she said.

Her final essay tells how she came to embrace her natural hair . She wrote about going to a mostly white grade school where classmates made jokes about her afro. When her grandmother sent her back with braids or cornrows, they made fun of those too.

Over time, she ignored their insults and found beauty in the styles worn by women in her life. She now runs a business doing braids and other hairstyles in her neighborhood.

“I stopped seeing myself through the lens of the European traditional beauty standards and started seeing myself through the lens that I created,” Amofa wrote.

“Criticism will persist, but it loses its power when you know there’s a crown on your head!”

Ma reported from Portland, Oregon.

The Associated Press’ education coverage receives financial support from multiple private foundations. AP is solely responsible for all content. Find AP’s standards for working with philanthropies, a list of supporters and funded coverage areas at AP.org .

Copyright © 2024 The Associated Press. All rights reserved. This material may not be published, broadcast, written or redistributed.

Related News

Trader Joe’s upped the price of its bananas for the first time in decades. Here’s why

Looking at a solar eclipse can be dangerous without eclipse glasses. Here’s what to know

Recommended

2 bodies recovered from site of Baltimore bridge collapse

DC reaches deal to keep Capitals, Wizards at Capital One Arena until 2050

What we know about the missing workers presumed dead after Baltimore bridge collapse

Related categories:.

More From Forbes

Prompt engineering embraces new essay-compression technique that expands on getting very constructive generative ai results.

• Share to Facebook
• Share to Twitter
• Share to Linkedin

Handy new prompt engineering technique to do compression on essays and other text is examined and ... [+] showcased as being quite useful.

In today’s column, I am continuing my ongoing coverage of prompt engineering strategies and tactics that aid in getting the most out of using generative AI apps such as ChatGPT, GPT-4, Bard, Gemini, Claude, etc. The focus here is on a brand-new prompting approach that enables you to cleverly compress essays and other text-based narratives. This is a handy technique that ought to be in your prompting best-practices toolkit.

For my in-depth comprehensive guide on over thirty other keystone prompting strategies, see the link here .

Here’s the essence of today’s discussion.

One of the most popular uses of generative AI entails creating summaries of entered text, see my prior coverage on prompting strategies for summarization at the link here and also the chain-of-density technique (CoD) at the link here . The deal is this. You might have a lengthy document that needs a helpful summary covering the crucial points so that you don’t have to wade through a convoluted textual morass. The simplest prompt to achieve a summary will merely tell the AI to go ahead and summarize the material. Boom, you are done, or so it seems.

The problem is that the summary might not be very good. Odds are that the generative AI might omit points that a human would have realized are vital to include in a summary. There is also a significant chance that the source material will be interpreted rather than carried straight ahead into the summary. Loosey-goosey interpretations of what was said in the source could be way off base.

Into this realm comes the act of compression.

Rather than producing a summary that might veer from the source, you can ask to do a compression. The idea of compression is that the source is reduced in size by a thinning process, but no adaptations or rephrasing takes place. You are still seeing the same words that were in the source. A thinning process is undertaken to try and keep just the needed meat on the bones conveying the crux of the source material.

Best High-Yield Savings Accounts Of 2024

Best 5% interest savings accounts of 2024.

To clarify, compression is not a silver bullet. Issues about compression can arise. For example, it is conceivable that as generative AI thins out a source text, some key points might inadvertently get weaned out. Another possible issue is that the resulting compressed text might seem nearly unreadable or incomprehensible. The words removed that might have seemed inconsequential by the AI were possibly integral to the intelligible reading of the text.

The bottom line is that anyone proficient in the use of generative AI should know both techniques by heart, freely welding the best ways to do prompting that encompasses summarization and astute prompting that does compression. You can then wisely decide which approach befits a given circumstance. Sometimes you pull out the hammer of summarization, and other times you reach for the screwdriver of compression. They are two distinct tools that require suitable handling and must be selected for the appropriate situations.

A frequent question I get during the prompt engineering classes that I teach is whether summarization is better than compression, or whether compression is better than summarization. This always brings a smile to my face. The conveyed conundrum is a false choice. None is especially better than the other in the abstract. Each provides a set of benefits and drawbacks. I don’t think you can in the abstract make a solid case that a hammer is better than or worse than a screwdriver since the situation at hand dictates as such. The same goes for summarization versus compression.

Okay, now that we’ve got that foundation established, there are already various recommended ways to word a prompt to do summarizations and compressions. A recent research study explored the topic and proffered a prompt for compression that they extensively tested and found to be quite useful. I will walk you through their research efforts and results.

I opted to try out the compression prompt by doing some of my own testing as a mini-experiment using ChatGPT and GPT-4. This allows you to see the compression prompting in action. I will discuss the matter and aim to arm you with how to suitably use the clever technique.

Before we get into the specifics of this new compression technique, it would be useful to make sure we are all on the same page about the nature and importance of prompt engineering.

Let’s do that.

The Nature And Importance Of Prompt Engineering

Please be aware that composing well-devised prompts is essential to getting robust results from generative AI and large language models (LLMs). It is highly recommended that anyone avidly using generative AI should learn about and regularly practice the fine art and science of devising sound prompts. I purposefully note that prompting is both art and science. Some people are wanton in their prompting, which is not going to get you productive responses. You want to be systematic leverage the science of prompting, and include a suitable dash of artistry, combining to get you the most desirable results.

My golden rule about generative AI is this:

• The use of generative AI can altogether succeed or fail based on the prompt that you enter.

If you provide a prompt that is poorly composed, the odds are that the generative AI will wander all over the map and you won’t get anything demonstrative related to your inquiry. Similarly, if you put distracting words into your prompt, the odds are that the generative AI will pursue an unintended line of consideration. For example, if you include words that suggest levity, there is a solid chance that the generative AI will seemingly go into a humorous mode and no longer emit serious answers to your questions.

Be direct, be obvious, and avoid distractive wording.

Being copiously specific should also be cautiously employed. You see, being painstakingly specific can be off-putting due to giving too much information. Amidst all the details, there is a chance that the generative AI will either get lost in the weeds or will strike upon a particular word or phrase that causes a wild leap into some tangential realm. I am not saying that you should never use detailed prompts. That’s silly. I am saying that you should use detailed prompts in sensible ways, such as telling the generative AI that you are going to include copious details and forewarn the AI accordingly.

You need to compose your prompts in relatively straightforward language and be abundantly clear about what you are asking or what you are telling the generative AI to do.

A wide variety of cheat sheets and training courses for suitable ways to compose and utilize prompts has been rapidly entering the marketplace to try and help people leverage generative AI soundly. In addition, add-ons to generative AI have been devised to aid you when trying to come up with prudent prompts, see my coverage at the link here .

AI Ethics and AI Law also stridently enter into the prompt engineering domain. For example, whatever prompt you opt to compose can directly or inadvertently elicit or foster the potential of generative AI to produce essays and interactions that imbue untoward biases, errors, falsehoods, glitches, and even so-called AI hallucinations (I do not favor the catchphrase of AI hallucinations, though it has admittedly tremendous stickiness in the media; here’s my take on AI hallucinations at the link here ).

There is also a marked chance that we will ultimately see lawmakers come to the fore on these matters, possibly devising and putting in place new laws or regulations to try and scope and curtail misuses of generative AI. Regarding prompt engineering, there are likely going to be heated debates over putting boundaries around the kinds of prompts you can use. This might include requiring AI makers to filter and prevent certain presumed inappropriate or unsuitable prompts, a cringe-worthy issue for some that borders on free speech considerations. For my ongoing coverage of these types of AI Ethics and AI Law issues, see the link here and the link here , just to name a few.

All in all, be mindful of how you compose your prompts.

By being careful and thoughtful you will hopefully minimize the possibility of wasting your time and effort. There is also the matter of cost. If you are paying to use a generative AI app, the usage is sometimes based on how much computational activity is required to fulfill your prompt request or instruction. Thus, entering prompts that are off-target could cause the generative AI to take excessive computational resources to respond. You end up paying for stuff that either took longer than required or that doesn’t satisfy your request and you are stuck for the bill anyway.

I like to say at my speaking engagements that prompts and dealing with generative AI is like a box of chocolates. You never know exactly what you are going to get when you enter prompts. The generative AI is devised with a probabilistic and statistical underpinning which pretty much guarantees that the output produced will vary each time. In the parlance of the AI field, we say that generative AI is considered non-deterministic.

My point is that, unlike other apps or systems that you might use, you cannot fully predict what will come out of generative AI when inputting a particular prompt. You must remain flexible. You must always be on your toes. Do not fall into the mental laziness of assuming that the generative AI output will always be correct or apt to your query. It won’t be.

Write that down on a handy snip of paper and tape it onto your laptop or desktop screen.

Using Special Prompts Including Chain-of-Density

There is a slew of somewhat remarkable prompt phrases that are essential for anyone seriously doing prompt engineering. One such phrase involves telling generative AI to work on a stepwise basis, something commonly known as invoking chain-of-thought responses by the AI, see my coverage at the link here . Another popular ploy entails telling the AI to take a deep breath, see my analysis of this prompt at the link here . One of the latest favorites involves commanding the AI to take on a Star Trek consideration when devising an answer, see my discussion at the link here .

I previously covered in my writings the various prompts that are particularly suited for getting summaries of text that you might provide to generative AI. Summarizing text is a very common use of generative AI. You can simply paste a whole bunch of text into generative AI and get a pretty good summary of the text. That being said, summaries can diverge from the original content and potentially contain misleading or even outright incorrect summarized content, see my analysis at the link here .

In short, summarizing can be risky due to the generative AI opting to rephrase the original content. The rephrasing might fail to depict the proper meaning and intention of the source content. Envision using summarization on doctor’s notes via generative AI, which is then handed to a different physician and they only have the summary at their purview, see my exploration of these matters at the link here .

Thus, a conventional summarization could be disturbingly problematic.

There is a method of devising summaries that use a prompting strategy that aims to bolster generative AI toward attaining especially superb or at least better than usual kinds of summaries. The technique is known as chain-of-density (CoD). Anybody versed in prompt engineering ought to become familiar with this insightful technique. Consider chain-of-density as not only helpful for producing summaries but there are a lot of other benefits garnered by understanding how the technique works and how this can power up your overall prompting prowess all-told.

For my detailed coverage of the chain-of-density prompting technique, see the link here .

Allow me a moment to share some of those elicitations with you here.

When you are trying to craft a summary, you often might do so in a series of successive attempts. Your first shot might be to craft a summary that has only a few of the biggest points that need to be included. After considering the initial draft, the odds are that you might further refine the summary by adding more elements to it. This can go on and on. Depending on how thorough you are, you might do a handful or more of these refining iterations. Each iteration can be construed as a chain of iterative summaries, one leading to the next for a given instance of trying to write a summary.

That’s the “chain” part of this process.

Let’s add some further terminology to describe the summary-making effort.

A summary typically starts as somewhat sparse when you first toss it together. There isn’t much of any substance in the summary. You are usually seeking to further pack substance into the summary and do so while fighting the length of the summary. The more substance that you can jam into the summary, the higher the density of the summary.

We can give a name to the substance by saying that we are trying to identify important “entities” within the original content. Those entities might be facts or figures. The entities are said to be anything especially instrumental to the overall meaning of the original content. A hope is to carry over as many of the demonstrative entities as feasible into the summary.

Your summary-making process then is to iteratively devise a summary by starting with a sparse version and then adding more and more entities or substances to increase the density until you reach some desired or suitable end-state. The series of iterations acts as a chain. Each is used to connect to the next. You usually will retain the entities from one version to the next version, and be decidedly adding more of the entities available in the original as you seek to jampack the summary accordingly.

Reflect on the adage of putting five pounds of rocks into a three-pound bag.

Maybe you put one pound of rocks into the three-pound bag at the initial attempt. The bag is considered sparsely populated. There is still room to spare. The density is low. You then put a second pound of rocks into the bag. The density is increasing. The sparseness is lessening. Finally, you put in a third pound of rocks. You have hit the maximum density and the sparseness has presumably dropped to near zero.

Suppose that the bag can be elongated.

Wonderful, you exclaim, being overjoyed at having more available space. Imagine though that you are going to hand the bag over to someone else. The larger and heavier the bag, the less useful it becomes. The same applies to summaries.

A rule of thumb is that you want to minimize the length or size of the summary, meanwhile maximizing the summarization content. The two factors are often in contention with each other. You are tempted to increase the length to get more substance included. The length being increased will potentially undercut that the summary is supposed to be a summary.

A person might seemingly just go ahead and read the original content if the summary approaches the size of the original material being summarized. The summary isn’t especially a summary anymore at that juncture. Indeed, sometimes a summary turns out to be longer than the original content that is supposedly being summarized.

How can this be, you might be thinking?

The answer has to do with being extractive versus being abstractive.

During the summarization process, you are looking at two possibilities of the content being carried over into the summary. First, you aim to be extractive, primarily extracting key aspects and shoveling those into the summary. Second, you might at times be abstractive, whereby you go beyond the words themselves of the original content and begin to reinterpret or perhaps elaborate beyond what the summary per se has to say.

A purely extractive summary is more likely to be construed as a fair and balanced reflection of the original content. You are not changing things up. You are only carrying the essentials (entities) over into the summary. The problem with an abstractive summary is that you are potentially changing up things and will be biasing or in some manner altering the meaning found within the original content being summarized. The danger is that this kind of summary is no longer seen as fair and balanced, and instead is based on the perceptions and opinions of the summarizer.

In a sense, if you want an unadorned straightforward summary, you are better off with an extractive summary. If you want an adorned or embellished summary, that goes beyond the words presented in the original source, you might seek an abstractive summary. The thing is, the abstractive summary might no longer be an apt reflection of the original source. That is also how the summary might become longer than the original since the embellishments can possibly increase the size of things and you could find yourself looking at a summary that is much longer than the source used for the summary.

Focusing On Compression As An Extractive Form Of Summarization

The usual parlance when you want a strictly extractive form of summarization is to say that you want a compression of your text.

I should warn you though that the word “compression” means different things to different people. Some are willing to be abstractive when doing a compression. Others would be aghast at someone veering into an abstractive mode when doing what they consider to be a rightfully done compression. To them, compression must be stridently adherent to not rewording any aspect of the source text being used.

I will be showing you in the next section some examples of using prompts related to asking generative AI to do compression. You will immediately see that the word “compression” is relatively ambiguous to most generative AI apps. They might interpret the word to imply purely extractive, but more likely they will assume this means to be abstractive. You will need to nail down in detail what you want the AI to do. I will show you this.

What else should you be thinking about when it comes to doing compression in generative AI?

I am glad you asked.

A recent research study entitled “LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression” by Zhuoshi Pan, QianhuiWu, Huiqiang Jiang, Menglin Xia, Xufang Luo, Jue Zhang, Qingwei Lin, Victor Rühle, Yuqing Yang, Chin-Yew Lin, H. Vicky Zhao, Lili Qiu, Dongmei Zhang, arXiv , March 19, 2024, made these salient points (excerpts):

• “This paper focuses on task-agnostic prompt compression for better generalizability and efficiency.”
• “We formulate prompt compression as a token classification problem to guarantee the faithfulness of the compressed prompt to the original one, and use a Transformer encoder as the base architecture to capture all essential information for prompt compression from the full bidirectional context.”
• “Recent years have witnessed the emergence of various prompting techniques for large language models (LLMs), such as Chain-of-Thought (COT), In-context Learning (ICL), and Retrieval Augmented Generation (RAG). These techniques empower LLMs to handle complex and varied tasks through rich and informative prompts that may exceed tens of thousands of tokens.”
• “However, the benefits of such lengthy prompts come at a cost of increased computational and financial overhead, as well as the degraded information perception ability of LLMs. Prompt compression is a straightforward solution to address these issues, which attempts to shorten the original prompts without losing essential information.”

I’d like to explain some of those notable points to you.

First, one ongoing consideration is whether a compression prompt is written to be task-specific or task-agnostic.

A task-specific prompt is worded to fit a particular task at hand, such as if you were doing a compression of medical notes and you wanted the prompt to emphasize particular characteristics associated with medical lingo. The prompt is devised for that domain.

A task-agnostic prompt is broader and allows you to use it in a wide variety of circumstances. It is always best to have in your tool chest a task-agnostic prompt since you can use it on a generalized basis, plus you can tweak the wording for task-specific settings. The aim of the above research study was to focus on task-agnostic prompts for compression, which is handy for our purposes herein.

Second, we often speak of compression prompts as hopefully being faithful to the source material.

Faithfulness refers to the notion that the compression tries to retain crucial wording and not goof by removing something that will substantively undercut the meaning. Imagine that a sentence said “Do not cook for more than two minutes” and the compression lopped out the word “not”, leading to this compressed version of “Do cook for more than two minutes”. Ouch, your souffle might be ruined by your having used an unfaithful compression.

Third, an aspirational aim of compression is to be bidirectional, allowing that besides going from the source to a proper compressed version, you potentially can use the compressed version to get back to the source. This is not necessarily the case and opens quite a can of worms. Not everyone needs or expects to be able to return to the full source.

Fourth, it is worthwhile to note frequent circumstances whereby compression really shines.

Suppose that you are importing a ton of text into generative AI, see my explanation of importations prompting strategies at the link here . Or maybe you are using in-context modeling and relying on an external source of text, doing so via the use of RAG (retrieval augmented generation), see my discussion at the link here . And so on.

An important question that comes up is whether it is sensible to bring in the whole ball of wax or whether to try and somehow shorten what you need to bring in.

If you bring in a massive amount of text, you are fighting against the potential limits of generative AI. Those limits can severely degrade the results of the AI trying to cope with the text overload. You also have to consider potential costs. If you are paying for the use of generative AI, the longer-sized text is bound to be a costlier charge than if you could thin it down. Likewise, the servers used to compute the generative AI actions will consume more computational cycles and you will get a double whammy on cost.

I am sure that you immediately discern that a potential solution would be to compress the content, thus, you don’t have to confront the technical limits of the generative AI and might be able to keep your costs lower too. Whenever you are considering bringing into generative AI any large-scale body of text, please have alarm bells go off in your head that tell you to consider doing either compression or summarization. You will be glad that you considered the option.

Returning to the above research study facets, let’s see what they did and the results they found (excerpts):

• “We conduct extensive experiments and analysis on both in-domain (i.e., MeetingBank) and out-of-domain datasets (i.e., LongBench, ZeroScrolls, GSM8K, and Big Bench Hard).” (ibid).
• “Despite being small in size, our model shows significant performance gains over strong baselines and demonstrates robust generalization ability from GPT-3.5-Turbo to Mistral-7B.” (ibid).
• “Additionally, our model is 3x-6x faster than existing prompt compression methods, while accelerating the end-to-end latency by 1.6x-2.9x with compression ratios of 2x-5x.” (ibid).

You can see that they were able to attain some impressive results.

I especially applaud research on generative AI that includes multiple generative AI apps. I say this because each generative AI app differs from the other, ergo if a study exclusively uses only one generative AI app it is hard to claim that the same approach will work equally well on other generative AI apps. Here, it is good news for us that they tried different generative AI apps and essentially found similarly positive results.

I am sure you are eager to see an example of what their compression prompt was able to achieve.

Here is one such example from their study, in this case compressing a meeting transcript snippet from a database of such transcripts (excerpt):

• Original Text: “Item 15, report from City Manager Recommendation to adopt three resolutions. First, to join the Victory Pace program. Second, to join the California first program. And number three, consenting to the inclusion of certain properties within the jurisdiction in the California Hero program. It was emotion, motion, a second and public comment. CNN. Please cast your vote. Oh. Was your public comment? Yeah. Please come forward. I thank you, Mr. Mayor. Thank you. Members of the council. My name is Alex Mitchell. I represent the hero program. Just wanted to let you know that the hero program. Has been in California for the last three and a half years.”
• Compressed Text: “Item 15, City Manager Recommendation adopt three resolutions. Join Victory Pace program. Join California first program. Consent inclusion properties jurisdiction California Hero program. Emotion, motion, second, public comment. Cast vote. Public comment? Come forward. Alex Mitchell, represent Hero program. Hero program in California three half years.”

The original text has 109 words, while the compressed text has 46 words. The compression removed 63 words (that’s 109 words in total minus 46 words remaining = 63 words removed). You could say that roughly 58% of the source was removed (i.e., 58% of 109 is about 63). Well over half of the source has been removed.

Now comes the tough part.

Does the compressed version suitably capture the crux of the original text?

Has anything been removed that is essential and thus the compressed version has missed the boat?

I think you can likely see how hard a problem it is to do a proper compression if you try doing the same task yourself by hand. Here’s what I am saying. I want you to look at the original text and try to compress it. What would you remove? What would you leave in? How far can you go on this process, yet still ensure that the result has the needed points and is intelligible?

A twist involves settings where the original text itself is not necessarily that intelligible. A transcript is a great example of a thorny problem. The odds are that a transcript will contain filler words and utterances. Are they worthy of being retained or should those be removed? If you exhort it is obvious to remove the utterances, the issue there is that the tenor or tone of what was said could end up being removed. Perhaps that aspect is vital to comprehending the matter.

Besides deciding whether a compression has done a good job, you might also be wondering whether there is even more compression that can be performed.

Envision that we use a compression that does a good job and reduces the size of the source to half in size. But along comes a better approach that can do the same job and achieve a three-quarters reduction in size. Assuming that all else is equal, you would naturally seek to use the three-quarters reduction approach.

This can be tricky to compare. The greater the compression in terms of reducing the size, the more likely the odds that the compression isn’t going to be as intelligible or otherwise isn’t as apt. The criteria are usually battling with each other. The more reduction, the less of a good job in the sense of being usable and appropriate. The more of a good job you want, the less reduction you can usually achieve. Darned if you do, darned if you don’t.

The Compression Prompt As A Task-Agnostic Template

Congratulations, you have waded through the weeds to get to the crescendo.

Let’s see what the prompt template of their task-agnostic compression prompt consists of (excerpt):

• “System Prompt: You are an excellent linguist and very good at compressing passages into short expressions by removing unimportant words while retaining as much information as possible.”
• “User Prompt: Compress the given text to short expressions, such that you can reconstruct it as close as possible to the original.”
• “Unlike the usual text compression, I need you to comply with the 5 conditions below:”
• “1. You can ONLY remove unimportant words.”
• “2. Do not reorder the original words.”
• “3. Do not change the original words.”
• “4. Do not use abbreviations or emojis.”
• “5. Do not add new words or symbols.”

-“Compress the original aggressively by removing words only. Compress the original as short as you can, while retaining as much information as possible. If you understand, please compress the following text: {text to compress}”

I will be using that same prompt in a moment when I show you examples using ChatGPT and GPT-4.

Let’s briefly review the prompt and see what it contains.

First, the prompt tells the generative AI to invoke a persona. For my extensive coverage of personas in generative AI, see the link here and the link here . This is a powerful prompting technique.

Personas get the AI into a computational pattern-matching mode that can significantly aid whatever task you are trying to perform. In this instance, they are telling the AI that it is to pretend to be an excellent linguist and good at compressing passages of text. Nice touch.

Many users of generative AI fail to lean into personas when doing so would be especially advantageous. The use of a persona for amping up compression is a good move. Make sure to keep personas in your mind whenever using generative AI. You’ll be glad that you did.

Second, they provide five stipulated conditions for the compression. This is good since you want to provide as much guidance as you can when asking for a compression. I mentioned earlier that the word “compression” can be ambiguous. Spelling out the five stated conditions will give the AI important clues about what to do.

Third, the phrasing for AI that you use in a prompt ought to at times be emphatic. If you aren’t emphatic, the instructions might be ignored or downplayed. For example, telling the generative AI to aggressively do the compression is a means of guiding the AI toward doing a strident job. I should note that your being emphatic has a downside. The emphatic wording, if it goes way over the top, can trigger a mathematical and computational pattern-matching storm that will distract from what you are trying to undertake.

I believe that sufficiently outlines the keystones of the prompt. Just like buying a new car, we need to take this shiny new vehicle for a drive. Get yourself strapped in and ready for a wild ride.

Showcasing The Topic Via Examples In Generative AI

I will next proceed to showcase what happens when you use compression prompting while in a conversation with generative AI. I am going to use ChatGPT and GPT-4 to showcase my examples. ChatGPT and GPT-4 are sensible choices in this case due to their immense popularity as generative AI apps. An estimated one hundred million weekly active users are said to be utilizing ChatGPT and GPT-4. That’s a staggering number.

A few quick comments before we launch into using ChatGPT and GPT-4.

If you are going to try to do the same prompts that I show here, realize that the probabilistic and statistical properties will likely produce slightly different results than what I show here. That’s the nature of generative AI and how it is devised.

You also should expect that different generative AI apps will respond in different ways. Not all generative AI apps are the same. The key is that sometimes a particular prompt will work in one generative AI app and not another. You will need to undertake a series of trial-and-error attempts to ferret this out.

There is also the timing factor. A prompt that worked today might not work the same way tomorrow. Most of the generative AI apps are continually being updated. The updates might alter internal facets that could change how the AI reacts to your prompting.

We are ready to get underway and I will begin with ChatGPT.

You can do the same in whichever generative AI app that you are using.

As a heads-up, I decided to use Abraham Lincoln’s famous Gettysburg Address as the text that I wanted to have compressed. There are variations of his speech that can be readily found online. I opted to use a version that contains 271 words. Our aim in this mini-experiment will be to do a compression and see if we can reduce the size, along with retaining intelligibility and vital points.

I’m sure that you already know the speech, which starts this way “Four score and seven years ago our fathers brought forth on this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal. Now we are engaged in a great civil war, testing whether that nation, or any nation so conceived and so dedicated, can long endure…”

Here we go.

• My entered prompt into ChatGPT: “I want you to compress an essay that I am going to give you. Are you ready?”
• ChatGPT generated response: “Yes, I'm ready to help compress your essay. Please provide the essay, and I'll assist you in condensing it.”
• My entered prompt into ChatGPT: “Here is the text of the Gettysburg address. Go ahead and compress it.”
• ChatGPT generated response: “Abraham Lincoln's Gettysburg Address, delivered during the American Civil War, reflects on the nation's founding principles of liberty and equality. He honors those who sacrificed their lives on the battlefield, emphasizing the ongoing responsibility to preserve freedom and uphold democratic government. Lincoln urges dedication to the unfinished task of ensuring the nation's survival and the continuation of its principles, emphasizing the importance of honoring the fallen and ensuring their sacrifice was not in vain.”

Take a look at the so-called compression that I got.

You right away certainly observed that the generated response is not a compression. It is more of a summary. Well, maybe not quite a summary either. It has indications that go beyond the text that was provided. This likely was tapped into by ChatGPT as per other prior data training about Abraham Lincoln and the Gettysburg Address in general during the initial setup of ChatGPT.

I’m glad this happened. You see, I wanted to show you this kind of an example to emphasize my earlier point that the word “compression” can be ambiguous. This was not the kind of compression I had in mind.

If at first, you don’t succeed, try and try again, as they say.

• My entered prompt into ChatGPT: “Let’s start over. I want you to compress the material. You are to use only the material provided and compress the contents. Don’t make any interpretations. Don’t do any rewording.”
• ChatGPT generated response: “Understood. Please provide the material you'd like me to compress, and I'll follow your instructions accordingly.”
• My entered prompt into ChatGPT: ”Here is the text to be compressed…”
• ChatGPT generated response: "87 years ago, our fathers founded a nation dedicated to liberty and equality. Now, in a great civil war, we honor those who sacrificed for our nation. It's fitting to dedicate this ground, yet their deeds surpass our words. We must continue their work, ensuring their sacrifices were not in vain. Let us resolve to uphold freedom and democratic government.”

I valiantly tried to convey to ChatGPT that I wanted a compression.

Though the result is more so a compression than the first run, it still isn’t a compression of the caliber we are looking for. One irksome thing was the “87 years ago” which is a kind of summary or interpretation of what the source says. Didn’t want that.

But I am somewhat pleased that the result came out badly because it shows that even if you try to be instructive about compression, the result still might not come out as hoped for. I think we are primed and ready to see if the compression prompting template will do any better with ChatGPT.

Are you ready to see?

Hold your breath.

• My entered prompt (based on the research prompt template): “You are an excellent linguist and very good at compressing passages into short expressions by removing unimportant words while retaining as much information as possible. Compress the text I am going to give you and compress the text into short expressions, and such that you (ChatGPT) can reconstruct it as close as possible to the original. Unlike the usual text compression, I need you to comply with 5 conditions. First condition: You can ONLY remove unimportant words. Second condition: Do not reorder the original words. Third condition: Do not change the original words. Fourth condition: Do not use abbreviations or emojis. Fifth condition: Do not add new words or symbols. Make sure to compress the original text aggressively by removing words only. Compress the original text as short as you can, while retaining as much information as possible. Do you understand these instructions?”
• ChatGPT generated response : “Yes, I understand the instructions. Please provide the text you'd like me to compress.”
• My entered prompt: ”Here is the text to be compressed…”
• ChatGPT generated response: “Four score and seven years ago our fathers brought forth on this continent a new nation, conceived in Liberty, dedicated to the proposition that all men are created equal. Now engaged in a great civil war, testing whether that nation, or any nation so conceived and dedicated, can long endure. We are met on a great battle-field. We have come to dedicate a portion of that field, a final resting place for those who gave their lives. It is fitting and proper that we should do this. But, in a larger sense, we cannot dedicate, consecrate, hallow this ground. The brave men, living and dead, who struggled here, consecrated it. The world will little note, nor long remember what we say here, but it can never forget what they did here. It is for us the living to be dedicated to the unfinished work they who fought here have nobly advanced. It is for us to be dedicated to the great task remaining before us—that from these honored dead we take increased devotion to that cause—that we highly resolve that these dead shall not have died in vain—that this nation, under God, shall have a new birth of freedom—and that government of the people, by the people, for the people, shall not perish.”

Finally, we got a compression.

We are now in the right ballpark.

The compressed result has 217 words.

The original text has 271 words, while the compressed text has 217 words. The compression removed 54 words (that’s 271 words in total minus 217 words remaining = 54 words removed). You could say that roughly 20% of the source was removed (i.e., 20% of 271 is about 54). Approximately one-fifth was removed.

I dare say that’s not a compression ratio to be overly elated about. Look, I’m glad that we got the 20%, and I don’t like to look a gift horse in the mouth. It would seem intuitively we might be able to do better.

Note that the prompt doesn’t give a target level of compression. This is something you might consider adding to the prompt. I would first start without a target and just see what you get. Once you’ve seen the result of a first run, go ahead and consider redoing the prompt with a desired target level.

We must also of course consider whether the compressed version is any good at the get-go.

In the case of something like the Gettysburg Address, we could argue forever about the portions that were removed. It is tough since the entire speech is remarkable. Nearly every line and each word is astounding. Imagine trying to compress a poem, and the odds are that you will be leaving out facets that will break your heart.

Another reminder is that you should carefully consider what text you are going to compress and whether the compression, no matter how it works, will be able to reasonably do justice to the original text.

I went ahead and played with the compression prompt template and tried all kinds of variations. In the end, I decided that ChatGPT just doesn’t have the gravitas to go much further on this. I became exasperated and thought that for sure GPT-4, being more advanced than ChatGPT, would do a better job.

Let’s try using GPT-4 and see what happens.

• ​​ My entered prompt into GPT-4: “I want you to compress an essay that I am going to give you. Here is the text to be compressed…”
• GPT-4 generated response:

“To compress the text while retaining its essence, I'll remove some parts and rephrase others for brevity.”

"Four score and seven years ago, our fathers founded a new nation on this continent, dedicated to liberty and the belief that all men are created equal. We are now in the midst of a great civil war, testing whether this nation, or any nation so conceived, can endure. We meet on a battle-field of that war, dedicating a part of it as a final resting place for those who gave their lives for this nation's survival. It's fitting and proper we do this.

“Yet, we can't dedicate, consecrate, or hallow this ground beyond what the brave men, living and dead, who struggled here have done. The world may not remember our words, but it can't forget their deeds. It's our task to be dedicated to the unfinished work they've advanced so nobly. From these honored dead we take increased devotion to their cause, resolving that these dead shall not have died in vain, that this nation shall have a new birth of freedom, and that government of, by, and for the people shall not perish."

“The compressed text has been reduced to 176 words.”

End of GPT-4 response

You can see that I first tried a vague prompt about compression.

Sure enough, the compression included rephrasing, which isn’t what I had in mind. No problem. You can’t expect generative AI to be reading minds (well, you soon can, as I explain is being researched, see my coverage at the link here ).

Similar to what I had done with ChatGPT, I will next try a compression prompt based on some simple wording. I will also include a target level of compression so you can see how that impacts the result.

Here’s what took place.

• My entered prompt into GPT-4: “Do the compression over again. You cannot do any rephrasing this time. You can only remove words. Compress as much as possible while abiding by these rules. Try to aim for a result that has 100 words or less.”

“To compress the text by only removing words without rephrasing, I'll select key sentences and phrases that capture the essence of the original message while significantly reducing the length. Given the constraints, this selection process aims to preserve the core themes of unity, dedication, and the importance of the struggle while adhering to the word count limit.”

"Four score and seven years ago our fathers brought forth on this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal. Now we are engaged in a great civil war, testing whether that nation, or any nation so conceived and so dedicated, can long endure. The brave men, living and dead, who struggled here, have consecrated it, far above our poor power to add or detract. It is for us the living, to be dedicated to the unfinished work which they who fought here have thus far so nobly advanced."

“The condensed version of the text is approximately 99 words in size while adhering to the rule of only removing words without rephrasing.”

Seems like GPT-4 did what I asked.

As always, it is important to gauge the intelligibility and aptness of the compression. I won’t delve into that here, other than to say it seemed basically on par with what might be expected, especially given the severe constraint of being around 100 words or less.

We are ready to try the compression prompt template.

• My entered prompt into GPT-4 (based on the research prompt template): “You are an excellent linguist and very good at compressing passages into short expressions by removing unimportant words while retaining as much information as possible. Compress the text I am going to give you and compress the text into short expressions, and such that you (ChatGPT) can reconstruct it as close as possible to the original. Unlike the usual text compression, I need you to comply with 5 conditions. First condition: You can ONLY remove unimportant words. Second condition: Do not reorder the original words. Third condition: Do not change the original words. Fourth condition: Do not use abbreviations or emojis. Fifth condition: Do not add new words or symbols. Make sure to compress the original text aggressively by removing words only. Compress the original text as short as you can, while retaining as much information as possible. Do you understand these instructions?”
• GPT-4 generated response : “Yes, I understand the instructions. Please provide the text you'd like me to compress.”
• My entered prompt into GPT4: ”Here is the text to be compressed…”
• GPT-4 generated response: “Four score seven years ago fathers brought forth continent, new nation, conceived Liberty, dedicated proposition men created equal. Now engaged great civil war, testing whether nation, any nation conceived dedicated, can long endure. Met great battle-field war. Come dedicate portion field, final resting place who gave lives nation might live. Altogether fitting proper do this. But, larger sense, can not dedicate—not consecrate—not hallow—ground. Brave men, living dead, struggled, have consecrated, above our power add detract. World will little note, nor long remember say here, but never forget did here. For us living, rather, dedicated here unfinished work fought have nobly advanced. Rather for us here dedicated great task remaining before us—that from these honored dead take increased devotion cause gave last full measure devotion—that here highly resolve dead not died in vain—that nation, under God, shall new birth freedom—and government people, by people, for people, shall not perish earth.”

Note that I didn’t give a target level this time and thus the size of the result was based on the general tenor of the compression prompt about being aggressive on the compression.

The original text has 271 words, while the compressed text has 156 words. The compression removed 115 words (that’s 271 words in total minus 156 words remaining = 115 words removed). You could say that roughly 42% of the source was removed (i.e., 42% of 271 is about 115). We were nearing maybe half being removed. Remarkable.

I’ll mention that as always, we would want to closely scrutinize the compressed version. All the other caveats I’ve mentioned still apply.

My recommendation to you is that you try doing some compression prompting exercises via whichever generative AI app you are customarily using. Do so now, before you suddenly discover the need to do a compression. It would be wise to be prepared.

Proceed as follows. First, make use of vague compression prompts and see what happens. Next, fine-tune those prompts and look anew at what happens. Then, try the mentioned compression prompt and gauge how it goes for you. Tweak the prompt as needed. Keep on practicing until you feel you’ve gotten the right swing of things. Reward yourself accordingly.

A final thought for now on this matter.

John Morley, 1st Viscount Morley of Blackburn, the famed British statesman, and journalist, said this about compression: “The essence of a quote is the compression of a mass of thought and observation into a single saying.”

Compression is not an easy task. Any wanton approach can easily do compression if you are willing to toss intelligibility and aptness out the window. Getting generative AI to do erstwhile compression requires soundness on the part of the person entering the prompts. They also need to carefully review the result and not wantonly hand off the compressed version to others who might be unaware of what the original had to say.

Please remember that I energetically asked you to consider practicing doing compressions in generative AI, so that when the day arrives that you must do so, you are ready with the axe firmly in hand.

The last line today suitably goes to Abraham Lincoln: "Give me six hours to chop down a tree and I will spend the first four sharpening the ax."

• Editorial Standards
• Reprints & Permissions

By the BOOK

Morgan Parker Says ‘Poetry Is Under Everything’ She Writes

Crafting the arguments in “You Get What You Pay For,” her first essay collection, “felt like pulling apart a long piece of taffy,” says the author of “Magical Negro.”

Credit... Rebecca Clarke

Supported by

• Share full article

What books are on your night stand?

The craft anthology “How We Do It,” edited by the great Jericho Brown, and Shayla Lawson’s astounding “How to Live Free in a Dangerous World.”

Describe your ideal reading experience (when, where, what, how).

Probably on the smoking patio of a wine bar at happy hour on a sunny day, with a pencil in my hand and Dorothy Ashby or Ambrose Akinmusire playing through noise-canceling headphones. Or just a quiet morning on my couch with coffee, so engrossed I forget to flip the record.

What’s the last book you read that made you laugh?

“Erasure,” by Percival Everett . I picked up a used copy at Shakespeare & Company recently — after seeing Cord Jefferson’s brilliant adaptation , “American Fiction” — and even on a reread, it made me laugh out loud from the first page.

The last book that made you cry?

Weird or obnoxious if I say my own? Before that, it was probably Y.A.

Do you count any books as guilty pleasures?

That category’s filled to the brim and beyond by reality TV.

How do you organize your books?

Loosely or not at all. This is much to the horror of my Virgo pals, and while I used to take pride in navigating my shelves on familiarity alone, it’s something I’ve vowed to work on. Still, I doubt I’ll ever be an alphabetical type, and clearly I find genre segregation constricting. I do group things thematically, or even interpersonally — music biographies, Black Panthers, Harlem Renaissance; Jessica Hopper is next to John Giorno, and Chase Berggrun’s “R E D” is next to “Dracula”; Julie Buntin’s “Marlena” is beside her husband Gabe Habash’s “Stephen Florida”; Alison C. Rollins is next to her partner Nate Marshall is next to his bestie José Olivarez. At some point Hilton Als’s “White Girls” ended up next to “Male Fantasies,” and I don’t think I’ll ever separate them.

Which genres do you avoid?

There’s an essay in “You Get What You Pay For” where I mention reading a self-help book (as recommended by my now-former psychiatrist). I’d never read one before and have not since.

How does your poetry relate to your essay writing?

The truth is that poetry is under everything. It’s the lyric and sensory backbone. It’s what drives the sound, pace and imagery. (Everyone knows the best prose writers write and read poetry.) But while a poem strives for precision of language, the essay strives for precision of thought, even argument. In a poem, you can build (or approximate) an argument by plopping two images next to each other. It persuades by pointing. Writing these essays felt like pulling apart a long piece of taffy — I found myself reiterating a lot of what I’ve already expressed in poems, so it almost became a project of stretching out each poetic line, breaking down each concept to its root. The process is about asking, pondering, searching — and letting language take part in the answering.

You have a knack for terrific book titles. How did you name your new collection?

Thank you! I love a good title, but I also acknowledge the high bar I have set for myself. With this one, I struggled a bit, I think because it took me a while to understand the book myself, let alone how to introduce it to the world. The essays encompass a lot of seemingly disparate themes and even tonal registers, so framing the overall collection was daunting. I’d been tossing around a couple of options, including “Cheaper Than Therapy,” which appears as an essay title, when Jay-Z made the choice for me. I was in Italy at a residency, grieving the recent loss of my aunt and watching the “Big Pimpin’” video over and over as I worked on an essay about it for the book. I’d left my heavily tabbed copy of “Decoded” at home in Los Angeles, but was scrolling a PDF for details about the video shoot when I came across the line: “If the price is life, then you better get what you paid for.”

You describe yourself as foolish for believing “words could be the pathway to empathy and writing an active resistance against hate.” Might publishing this book change your mind?

Honestly? It’s my only hope.

What’s the last book you recommended to a member of your family?

“Heavy,” by Kiese Laymon, to my mom; Blair LM Kelley’s “ Black Folk: The Roots of the Black Working Class,” to my dad; and “A Is for Activist,” to my 8-month-old cousin.

What do you plan to read next?

Phillip B. Williams’s “Ours” was just published, and I’ve been excited about it for literally years. Vinson Cunningham’s “Great Expectations” came out the same day as my book, so I plan to make that my tour read.

You’re organizing a literary dinner party. Which three writers, dead or alive, do you invite?

June Jordan, Zora Neale Hurston, James Baldwin — but I’d be lying if I said I wouldn’t get just as much fun and fulfillment from a night with Angel Nafis, Danez Smith and Saeed Jones.

Explore More in Books

Want to know about the best books to read and the latest news start here..

James McBride’s novel sold a million copies, and he isn’t sure how he feels about that, as he considers the critical and commercial success  of “The Heaven & Earth Grocery Store.”

How did gender become a scary word? Judith Butler, the theorist who got us talking about the subject , has answers.

You never know what’s going to go wrong in these graphic novels, where Circus tigers, giant spiders, shifting borders and motherhood all threaten to end life as we know it .

When the author Tommy Orange received an impassioned email from a teacher in the Bronx, he dropped everything to visit the students  who inspired it.

Do you want to be a better reader?   Here’s some helpful advice to show you how to get the most out of your literary endeavor .

Each week, top authors and critics join the Book Review’s podcast to talk about the latest news in the literary world. Listen here .

Advertisement

I Hope You All Feel Terrible Now

How the internet—and Stephen Colbert—hounded Kate Middleton into revealing her diagnosis

Updated at 4:04 p.m ET on March 22, 2024

For many years, the most-complained-about cover of the British satirical magazine Private Eye was the one it published in the week after the death of Diana, Princess of Wales, in 1997. At the time, many people in Britain were loudly revolted by the tabloid newspapers that had hounded Diana after her divorce from Charles, and by the paparazzi whose quest for profitable pictures of the princess ended in an underpass in Paris.

Under the headline “Media to Blame,” the Eye cover carried a photograph of a crowd outside Buckingham Palace, with three speech bubbles. The first was: “The papers are a disgrace.” The next two said: “Yeah, I couldn’t get one anywhere” and “Borrow mine, it’s got a picture of the car.” People were furious. Sacks of angry, defensive mail arrived for days afterward, and several outlets withdrew the magazine from sale. (I am an Eye contributor, and these events have passed into office legend.) But with the benefit of hindsight, the implication was accurate: Intruding on the private lives of the royals is close to a British tradition. We Britons might have the occasional fit of remorse, but that doesn’t stop us. And now, because of the internet, everyone else can join in too.

Read: Just asking questions about Kate Middleton

That cover instantly sprang to mind when, earlier today, the current Princess of Wales announced that she has cancer. In a video recorded on Wednesday in Windsor, the former Kate Middleton outlined her diagnosis in order to put an end to weeks of speculation, largely incubated online but amplified and echoed by mainstream media outlets, about the state of her health and marriage.

Kate has effectively been bullied into this statement, because the alternative—a wildfire of gossip and conspiracy theories—was worse. So please, let’s not immediately switch into maudlin recriminations about how this happened. It happened because people felt they had the right to know Kate’s private medical information. The culprits may include three staff members at the London hospital that treated her, who have been accused of accessing her medical records, perhaps driven by the same curiosity that has lit up my WhatsApp inbox for weeks. Everyone hates the tabloid papers, until they become them.

In her statement, Kate said that after her abdominal surgery earlier in the year, which the press was told at the time was “planned”—a word designed to minimize its seriousness—later tests revealed an unspecified cancer. She is now undergoing “preventative chemotherapy,” but has not revealed the progression of the disease, or her exact prognosis. “I am well,” she said, promising that she is getting stronger every day. “I hope you will understand that as a family, we now need some time, space and privacy while I complete my treatment.”

This news will surely make many people feel bad. The massive online guessing game about the reasons for Kate’s invisibility seems far less fun now. Stephen Colbert’s “spilling the tea” monologue , which declared open season on the princess’s marriage, should probably be quietly interred somewhere. The sad simplicity of today’s statement, filmed on a bench with Kate in casual jeans and a striped sweater, certainly gave me pause. She mentioned the difficulty of having to “process” the news, as well as explaining her condition to her three young children in terms they could understand. The reference to the importance of “having William by my side” was pointed, given how much of the speculation has gleefully dwelt on the possibility that she was leaving him or vice versa.

Read: The eternal scrutiny of Kate Middleton

However, the statement also reveals that the online commentators who suggested that the royal household was keeping something from the public weren’t entirely wrong. Kate’s condition was described as noncancerous when her break from public life was announced in late January . The updated diagnosis appears to have been delivered in February, around the time her husband, Prince William, abruptly pulled out of speaking at a memorial service for the former king of Greece. Today’s statement represents a failure of Kensington Palace to control the narrative: first, by publishing a photograph of Kate and her children that was so obviously edited that photo agencies retracted it, and second, by giving its implicit permission for the publication of a grainy video of the couple shopping in Windsor over the weekend. Neither of those decisions quenched the inferno raging online—in fact, they fed it.

Some will say that Kate has finally done what she should have done much earlier: directly address the rumors in an official video, rather than drip-feed images that raised more questions than they answered. King Charles III has taken a different approach to his own (also unspecified) cancer, allowing footage to be filmed of him working from home. But then again, Kate has cancer at 42, is having chemo, and has three young children. Do you really have it in you to grade her media strategy and find it wanting?

Ironically, Britain’s tabloid papers have shown remarkable restraint; as I wrote earlier this month , they declined to publish the first paparazzi pictures of Kate taken after her withdrawal from public life. They have weighted their decisions toward respect and dignity—more so than the Meghan stans, royal tea-spillers, and KateGate theorists, who have generated such an unstoppable wave of interest in this story that its final destination was a woman with cancer being forced to reveal her diagnosis. If you ever wanted proof that the “mainstream media” are less powerful than ever before, this video of Kate Middleton sitting on a bench is it.

• Newsletters
• Account Activating this button will toggle the display of additional content Account Sign out

Even the Supreme Court’s Conservatives Are Fed Up With the Garbage Coming Out of the 5th Circuit

What happens when a lawless judge and a terrible appeals court embrace the dopiest First Amendment claim you’ve ever heard out of pure spite toward a Democratic president? That would be Murthy v. Missouri , a brain-meltingly dumb case that the Supreme Court was unfortunate enough to hear oral arguments in on Monday. Murthy poses a question so asinine that to ask it is to answer it: Can government officials encourage social media companies to moderate certain content that they deem harmful—most importantly, disinformation about COVID-19 in the middle of the pandemic?

Yes, of course they can: The First Amendment does not gag public officials from urging Facebook or the Washington Post or anyone else to publish or not publish certain information, especially when it contains dangerous lies about a once-in-a-century pandemic that could exacerbate the crisis. The First Amendment bars government censorship, not government persuasion, and the Biden administration planted itself on the latter side of that bright line. At least six justices grasped this basic constitutional principle on Monday. Several of them used arguments to highlight how this inane case illustrates so much of what’s wrong with the judiciary today, and hinted at the dangers it could pose to American democracy in the future. That we should pay attention to. The rest was an unfortunate sideshow.

Like so many Supreme Court cases these days, Murthy is built atop a heap of fake facts. The case began when Missouri and Louisiana sued agencies and officials across the Biden administration, falsely accusing them of coercing social media companies into censoring their residents’ free speech. (These states later added a handful of fringe anti-vaxxers to the suit.) They filed their complaint in the Monroe Division of the Western District of Louisiana, where—surprise!—they were guaranteed to draw a Trump appointee, Terry Doughty, the one judge hearing cases in that division. Judge Doughty has a record of issuing nationwide injunctions against the Biden administration on the basis of dubious legal and factual analysis. Most notably, he issued a nationwide bar against Biden’s vaccine mandate for health care workers in an opinion riddled with anti-vax nonsense (which the Supreme Court reversed ).

Doughty, in other words, was certain to rule against Biden in the social media case. Even still, the opinion he handed down on July 4, 2023, was a humiliating mess of contradictions, fabrications , and (ironically) misinformation . Doughty adopted the plaintiffs’ theory that the administration “coerced” social media companies into removing “conservative” speech about COVID, including posts promoting hydroxychloroquine and rejecting the efficacy of vaccines. He accused government officials of launching a “coordinated campaign” to silence conservatives by forcing private companies to take down anti-vax content, as well as false claims about election fraud. And he issued one of the most sweeping injunctions in the history of the American legal system, prohibiting any employee—including Homeland Security, the State Department, the Department of Justice, and the FBI—from “engaging in any communication of any kind with social-media companies” encouraging content moderation.

Doughty, it turns out, grievously butchered the record to reach his conclusion. An exhaustive analysis by Mike Masnick proves that Doughty consistently misrepresented testimony and other evidence in the record to construct a conspiracy theory with zero basis in reality. He distorted emails and other exchanges to make them look coercive when they were nothing of the sort, cherry-picking and rearranging quotations to put them in a censorious light. Yet the hard-right U.S. Court of Appeals for the 5 th Circuit, where law goes to die , affirmed Doughty’s conclusions and upheld much of his injunction (while narrowing it in part). That move sent the administration racing to the Supreme Court for an emergency stay, which it granted in October, over the dissents of Justices Samuel Alito, Neil Gorsuch, and Clarence Thomas. Monday’s arguments confirmed that a majority is prepared to side with Biden on the merits.

The justices did not sound happy to have the case before them. Justice Sonia Sotomayor scolded Louisiana Solicitor General Benjamin Aguiñaga (a former Alito clerk) for his lack of candor. “I have such a problem with your brief,” she told him. “You omit information that changes the context of some of your claims. You attribute things to people who it didn’t happen to.” Aguiñaga was unable to defend Doughty’s more extreme deceptions. So he had to fall back on what Justice Elena Kagan called an “extremely expansive” argument: the notion that “encouraging people to suppress their own speech” violates the First Amendment, even if it isn’t coercive. Kagan drew upon her past work in the executive branch to explain that, actually, government officials do this all the time . “I’ve had some experience encouraging press to suppress their own speech,” Kagan told Aguiñaga. Example: “You just wrote a bad editorial. Here are the five reasons you shouldn’t write another one.” Is that really unconstitutional?

Justice Brett Kavanaugh, a former White House staff secretary, also provided a real-world perspective after Alito fumed that officials had been too mean to the platforms and would never treat the traditional press so sharply. Kavanaugh gave Alito a reality check: “I’d assumed, thought, [and] experienced government press people throughout the federal government who regularly call up the media and berate them,” he told Deputy Solicitor General Brian Fletcher, who defended the administration. “You said the anger was unusual. I wasn’t entirely clear on that from my own experience.” Translation: Government employees yell at members of the media all the time. The media can accept or reject their requests. That is how it works .

But did Biden officials berate social media employees? Only on a few rare occasions. Here’s what really happened: The platforms in question, most prominently Facebook and Twitter, invited the government to help them identify and remove the most toxic disinformation. At the time, as Fletcher explained, these companies were eager to present themselves as “partners” with the administration in its push for COVID vaccination. Officials could “flag” questionable content, but the platforms made their own judgment calls, without any demonstrable fear of retaliation. Everyone agrees that the government violates free speech when it explicitly coerces companies into censoring expression under the threat of punishment. But, Fletcher persuasively argued, no such thing happened here.

That left Aguiñaga to lean on the theory that the government contravenes the First Amendment when it merely “encourages” specific content moderation. Justice Amy Coney Barrett, a stickler for legible rules, sounded borderline aghast at this baggy, boundless standard; she lobbed a series of hypotheticals at Aguiñaga that showed he was pushing a meaningless and impracticable test. Chief Justice John Roberts tried to help Aguiñaga understand that multiple agencies with competing agendas may pressure platforms in contradictory directions; doesn’t that “dilute the concept of coercion significantly?” (Aguiñaga just sounded confused.) Kavanaugh pointed out that the plaintiffs’ position could prevent officials from urging the press to safeguard national security by, for instance, asserting that a forthcoming article might imperil the troops.

Even Gorsuch, who started out sympathetic to Aguiñaga, was exasperated by the end, complaining that Doughty’s “universal injunction” was part of the “epidemic” that had to be stopped. Only Alito and Thomas sounded like surefire votes for the plaintiffs after nearly two hours of arguments, and Thomas’ heart was not fully in it. Alito stood alone in his increasingly uncontrolled and seething aggrievement .

The pandemic may be over, but Murthy remains a hugely relevant case. Perhaps most obviously, it is a cautionary tale about the extremism of Trump judges like Doughty, who keep issuing these lawless injunctions at the behest of red-state politicians, and the 5 th Circuit, which keeps upholding them. A disproportionate number of the Supreme Court’s cases this term emerge from the 5 th Circuit, which keeps appeasing the Trump-appointed zealots attempting to seize unimaginable amounts of power from the democratic branches. The chief justice is attempting to crack down on the kind of judge-shopping that occurred here, but the fight is just beginning. The evident frustration of Roberts, Kavanaugh, Barrett, and even Gorsuch on Monday suggests that a majority of the court is fed up with this political manipulation of the judiciary by rogue judges with undisguised partisan loyalties.

So, yes, SCOTUS is likely to reach the right result in Murthy . Alarmingly, though, so much damage has already been done. In light of Doughty’s injunction, the federal government shut down all of its efforts to combat disinformation, fearful of judicial sanctions. Bad actors have exploited this development to flood the internet with lies about the upcoming 2024 election. If foreign entities attempt to interfere with the election via social media, the Biden administration will have few tools left to fight them. It is just too late to revive the various task forces that worked with platforms before Doughty disbanded them last year. For the Murthy plaintiffs, then, a Supreme Court defeat may not matter much. They have already done immense damage to truth and transparency on the internet. And at this late date, the worst of it cannot be undone.

Recommended

• Click to share on Facebook (Opens in new window)
• Click to share on Twitter (Opens in new window)
• Click to email a link to a friend (Opens in new window)
• Click to copy URL

Rebel Wilson says she lost her virginity at 35: ‘You shouldn’t feel pressure as a young person’

• View Author Archive
• Get author RSS feed

Thanks for contacting us. We've received your submission.

Rebel Wilson revealed that she lost her virginity at 35 and wants her experience to be a “positive message” to younger people.

“Not everybody has to lose their virginity as a teenager,” the “Pitch Perfect” star, now 44, told People ahead of the release of her new memoir, “Rebel Rising.”

“People can wait till they’re ready or wait till they’re a bit more mature,” she added. “And I think that could be a positive message. You obviously don’t have to wait until you’re in your thirties like me, but you shouldn’t feel pressure as a young person.”

Wilson said she used to avoid the topic of being a virgin as a teenager due to feeling “embarrassed.”

“There was one vague time, I think I told my best friend, ‘Oh, yeah, I just did it’ to just get it over with when I was like 23,'” she recalled. “Just to really avoid the questions.”

The “Senior Year” actress — who called herself a “late bloomer” — said she would sometimes “leave the room” when conversations about having sex would arise.

“And then the people that said, ‘Oh, at 24, it’s so late.’ And then I’m sitting here thinking, ‘Oh my god, my number’s 35. What the hell? I’m going to look like the biggest loser,'” she recounted.

Now that Wilson is in her 40s, she feels like she would’ve “explored” her sexuality more if she had been born 20 years later.

“I just knew I was attracted to men, and that was the normal thing,” she said.

“And so when I started opening myself up probably more after my father’s death and realizing, oh, even though I’d seen marriage as a terrible thing and waste of time, I started opening myself up to that. And then only years later, meeting women and having feelings for a woman, and that, I just think it’s a sign of where society kind of was.”

Wilson confirmed her engagement to her partner, Ramona Agruma , in February 2023.

Want more celebrity and pop culture news?

Start your day with Page Six Daily.

Thanks for signing up!

Please provide a valid email address.

By clicking above you agree to the Terms of Use and Privacy Policy .

Want celebrity news as it breaks? Hooked on Housewives?

“We said YES! 💗💗” she gushed in her Instagram caption . “Thank you @tiffanyandco for the stunning ring 💍 and to Bob Iger and the incredible team at Disneyland @disneyweddings for pulling off this magical surprise!”

Wilson also welcomed her first child , daughter Royce, via surrogate in November 2022.

“Rebel Rising” is out on April 2.

Share this article:

Advertisement

About iOS 17 Updates

iOS 17 brings big updates to Phone, Messages, and FaceTime that give you new ways to express yourself as you communicate. StandBy delivers a new full-screen experience with glanceable information designed to view from a distance when you turn iPhone on its side while charging. AirDrop makes it easier to share and connect with those around you and adds NameDrop for contact sharing. Enhancements to the keyboard make entering text faster and easier than ever before. iOS 17 also includes updates to Widgets, Safari, Music, AirPlay, and more.

For information on the security content of Apple software updates, please visit this website: https://support.apple.com/kb/HT201222

This update provides important bug fixes and security updates and is recommended for all users.

For information on the security content of Apple software updates, please visit this website:

https://support.apple.com/kb/HT201222

This update introduces new emoji, transcripts in Apple Podcasts and includes other features, bug fixes, and security updates for your iPhone.

New mushroom, phoenix, lime, broken chain, and shaking heads emoji are now available in the emoji keyboard

18 people and body emoji add the option to face them in either direction

Apple Podcasts

Transcripts let you follow an episode with text that highlights in sync with the audio in English, Spanish, French and German

Episode text can be read in full, searched for a word or phrase, tapped to play from a specific point and used with accessibility features such as Text Size, Increase Contrast, and VoiceOver

This update includes the following enhancements and bug fixes:

Music recognition lets you add songs you have identified to your Apple Music Playlists and Library, as well as Apple Music Classical

Siri has a new option to announce messages you receive in any supported language

Stolen Device Protection supports the option for increased security in all locations

Battery Health in Settings shows battery cycle count, manufacture date, and first use on iPhone 15 and iPhone 15 Pro models

Call Identification displays Apple-verified business name, logo, and department name when available

Business updates in Messages for Business provide trusted information for order status, flight notifications, fraud alerts or other transactions you opt into

Apple Cash virtual card numbers enable you to pay with Apple Cash at merchants that don’t yet accept Apple Pay by typing in your number from Wallet or using Safari AutoFill

Fixes an issue where contact pictures are blank in Find My

Fixes an issue for Dual SIM users where the phone number changes from primary to secondary and is visible to a group they have messaged

Some features may not be available for all regions or on all Apple devices. For information on the security content of Apple software updates, please visit this website:

This update provides bug fixes for your iPhone including:

Text may unexpectedly duplicate or overlap while typing

This update introduces additional security measures with Stolen Device Protection. This release also includes a new Unity wallpaper to honor Black history and culture in celebration of Black History Month, as well as other features, bug fixes, and security updates for your iPhone.

Stolen Device Protection

Stolen Device Protection increases security of iPhone and Apple ID by requiring Face ID or Touch ID with no passcode fallback to perform certain actions

Security Delay requires Face ID or Touch ID, an hour wait, and then an additional successful biometric authentication before sensitive operations like changing device passcode or Apple ID password can be performed

Lock Screen

New Unity wallpaper honors Black history and culture in celebration of Black History Month

Collaborate on playlists allows you to invite friends to join your playlist and everyone can add, reorder, and remove songs

Emoji reactions can be added to any track in a collaborative playlist

This update also includes the following improvements:

AirPlay hotel support lets you stream content directly to the TV in your room in select hotels

AppleCare & Warranty in Settings shows your coverage for all devices signed in with your Apple ID

Crash detection optimizations (all iPhone 14 and iPhone 15 models)

This update provides important bug fixes and is recommended for all users.

This update introduces Journal, an all-new way to reflect on life’s moments and preserve your memories. This release also includes Action button and Camera enhancements, as well as other features, bug fixes, and security updates for your iPhone.

Journal is a new app that lets you write about the small moments and big events in your life so you can practice gratitude and improve your wellbeing

Journaling suggestions make it easy to remember your experiences by intelligently grouping your outings, photos, workouts, and more into moments you can add to your journal

Filters let you quickly find bookmarked entries or show entries with attachments so you can revisit and reflect on key moments in your life

Scheduled notifications help you keep a consistent journaling practice by reminding you to write on the days and time you choose

Option to lock your journal using Touch ID or Face ID

iCloud sync keeps your journal entries safe and encrypted on iCloud

Action Button

Translate option for the Action button on iPhone 15 Pro and iPhone 15 Pro Max to quickly translate phrases or have a conversation with someone in another language

Spatial video lets you capture video on iPhone 15 Pro and iPhone 15 Pro Max so you can relive your memories in three dimensions on Apple Vision Pro

Improved Telephoto camera focusing speed when capturing small faraway objects on iPhone 15 Pro and iPhone 15 Pro Max

Catch-up arrow lets you easily jump to your first unread message in a conversation by tapping the arrow visible in the top-right corner

Add sticker option in the context menu lets you add a sticker directly to a bubble

Memoji updates include the ability to adjust the body shape of any Memoji

Contact Key Verification provides automatic alerts and Contact Verification Codes to help verify people facing extraordinary digital threats are messaging only with the people they intend

Precipitation amounts help you stay on top of rain and snow conditions for a given day over the next 10 days

New widgets let you choose from next-hour precipitation, daily forecast, sunrise and sunset times, and current conditions such as Air Quality, Feels Like, and wind speed

Wind map snapshot helps you quickly assess wind patterns and access the animated wind map overlay to prepare for forecasted wind conditions for the next 24 hours

Interactive moon calendar lets you easily visualize the phase of the moon on any day for the next month

This update also includes the following improvements and bug fixes:

Siri support for privately accessing and logging Health app data using your voice

AirDrop improvements including expanded contact sharing options and the ability to share boarding passes, movie tickets, and other eligible passes by bringing two iPhones together

Favorite Songs Playlist in Apple Music lets you quickly get back to the songs you mark as favorites

Use Listening History in Apple Music can be disabled in a Focus so music you listen to does not appear in Recently Played or influence your recommendations

A new Digital Clock Widget lets you quickly catch a glimpse of the time on your Home Screen and while in StandBy

Enhanced AutoFill identifies fields in PDFs and other forms enabling you to populate them with information such as names and addresses from your contacts

New keyboard layouts provide support for 8 Sámi languages

Sensitive Content Warning for stickers in Messages prevents you from being unexpectedly shown a sticker containing nudity

Qi2 charger support for all iPhone 13 models and iPhone 14 models

Fixes an issue that may prevent wireless charging in certain vehicles

This update provides important security fixes and is recommended for all users.

In rare circumstances, Apple Pay and other NFC features may become unavailable on iPhone 15 models after wireless charging in certain cars

Weather Lock Screen widget may not correctly display snow

This update introduces the ability for AirDrop transfers to continue over the internet when you step out of AirDrop range. This release also includes enhancements to StandBy and Apple Music, as well as other features, bug fixes, and security updates for your iPhone.

Content continues to transfer over the internet when you step out of AirDrop range

New options to control when the display turns off (iPhone 14 Pro, iPhone 14 Pro Max, iPhone 15 Pro, and iPhone 15 Pro Max)

Favorites expanded to include songs, albums, and playlists, and you can filter to display your favorites in the library

New cover art collection offers designs that change colors to reflect the music in your playlist

Song suggestions appear at the bottom of every playlist, making it easy to add music that matches the vibe of your playlist

Option to choose a specific album to use with Photo Shuffle on the Lock Screen

Home key support for Matter locks

Improved reliability of Screen Time settings syncing across devices

Fixes an issue that may cause the Significant Location privacy setting to reset when transferring an Apple Watch or pairing it for the first time

Resolves an issue where the names of incoming callers may not appear when you are on another call

Addresses an issue where custom and purchased ringtones may not appear as options for your text tone

Fixes an issue that may cause the keyboard to be less responsive

Fixes an issue that may cause display image persistence

https://support.apple.com/HT201222

This update provides important bug fixes, security updates, and addresses an issue that may cause iPhone to run warmer than expected.

This update provides important bug fixes, security updates, and fixes an issue that may prevent transferring data directly from another iPhone during setup.

Contact Posters let you customize how you appear on other people’s devices when you call them with a customized poster

Live Voicemail displays a live transcription as someone leaves a message and allows you to pick up the call

Stickers iMessage app brings all your stickers into one place including Live Stickers, Memoji, Animoji, emoji stickers, and your third party sticker packs

Live Stickers can be created by lifting the subject from photos or videos and stylizing them with effects like Shiny, Puffy, Comic, and Outline

Check In automatically notifies a family member or friend when you arrive at a destination safely and can share helpful information with them in case of a delay

Audio message transcription is available for audio messages you receive so you can read them in the moment and listen later

Search improvements help you find messages faster by allowing you to combine search filters such as people, keywords, and content types like photos or links to find exactly what you are looking for

Swipe to reply to a message inline by swiping to the right on any bubble

One-time verification code cleanup automatically deletes verification codes from the Messages app after using them with AutoFill in other apps

Leave a video or audio message to capture exactly what you want to say when someone does not pick up your FaceTime call

Enjoy FaceTime calls on Apple TV by using your iPhone as a camera (Apple TV 4K 2nd generation and later)

Reactions layer 3D effects like hearts, balloons, confetti, and more around you in video calls and can be triggered with gestures

Video effects allow you to adjust the intensity of Studio Lighting and Portrait mode

Full-screen experience with glanceable information like clocks, photos, and widgets designed to view from a distance when iPhone is on its side and charging in places such as your nightstand, kitchen counter, or desk

Clocks are available in a variety of styles including Digital, Analog, Solar, Float, and World Clock, with elements you can personalize like the accent color

Photos automatically shuffle through your best shots or showcase a specific album you choose

Widgets give you access to information at a distance and appear in Smart Stacks that deliver the right information at the right time

Night Mode lets clocks, photos, and widgets take on a red tone in low light

Preferred view per MagSafe charger remembers your preference for each place you charge with MagSafe, whether that’s a clock, photos, or widgets

Interactive widgets let you take actions, like mark a reminder as complete, directly from the widget by tapping it on the Home Screen, Lock Screen, or in StandBy

iPhone widgets on Mac enable you to add widgets from your iPhone to your Mac desktop

NameDrop lets you exchange contact information with someone new by bringing your iPhones close together

New way to initiate AirDrop allows you to share content or start a SharePlay session over AirDrop by bringing your iPhones close together

Improved autocorrect accuracy makes typing even easier by leveraging a powerful transformer-based language model (iPhone 12 and later)

Easier autocorrect editing temporarily underlines corrected words and lets you revert back to what you originally typed with just a tap

Enhanced sentence corrections can correct more types of grammatical mistakes when you finish sentences (iPhone 12 and later)

Inline predictive text shows single and multi-word predictions as you type that can be added by tapping space bar (iPhone 12 and later)

Safari and Passwords

Profiles keep your browsing separate for topics like work and personal, separating your history, cookies, extensions, Tab Groups, and favorites

Private Browsing enhancements include locking your private browsing windows when you’re not using them, blocking known trackers from loading, and removing identifying tracking from URLs

Password and passkey sharing lets you create a group of passwords to share with trusted contacts that stays up to date as members of the group make changes

One-time verification code AutoFill from Mail autofill in Safari so you can log in without leaving the browser

SharePlay makes it easy for everyone to control and play Apple Music in the car

Crossfade smoothly transitions between songs by fading out the currently playing song while fading in the next so the music never stops

Intelligent AirPlay device list makes finding the right AirPlay-compatible TV or speaker even easier by showing your devices in order of relevance, based on your preferences

Suggested AirPlay device connections are proactively shown to you as a notification to make it even more seamless to connect to your preferred AirPlay devices

Automatic AirPlay device connections are made between your iPhone and the most relevant AirPlay-compatible device so all you have to do is tap “Play” to begin enjoying your content

Adaptive Audio delivers a new listening mode that dynamically blends Active Noise Cancellation and Transparency to tailor the noise control experience based on the conditions of your environment (AirPods Pro (2nd generation) with firmware version 6A300 or later)

Personalized Volume adjusts the volume of your media in response to your environment and listening preferences over time (AirPods Pro (2nd generation) with firmware version 6A300 or later)

Conversation Awareness lowers your media volume and enhances the voices of the people in front of the user, all while reducing background noise (AirPods Pro (2nd generation) with firmware version 6A300 or later)

Press to mute and unmute your microphone by pressing the AirPods stem or the Digital Crown on AirPods Max when on a call (AirPods (3rd generation), AirPods Pro (1st and 2nd generation), or AirPods Max with firmware version 6A300 or later)

Offline Maps allow you to select an area you want to access, search, and explore rich information for places to download for use when your iPhone doesn’t have a Wi-Fi or cellular signal

EV routing improvements give you routes based on real-time EV charger availability for supported chargers

Option to say “Siri” in addition to “Hey Siri” for an even more natural way to make requests

Back-to-back requests can be issued without needing to reactivate Siri in between commands (iPhone 11 and later)

Visual Look Up

Expanded domains in Visual Look Up help you discover similar recipes from photos of food, Maps information from photos of storefronts, and the meaning of signs and symbols on things like laundry tags

Multiple or single subjects can be lifted from the background of photos and videos and placed into apps like Messages

Visual Look Up in Video helps you learn about objects that appear in paused video frames

Visual Look Up for subjects in photos enables you to look up information about objects you lift from photos directly from the callout bar

State of Mind reflection allows you to log your momentary emotion and daily mood, choose what factors are having the biggest impact on you, and describe your feelings

Interactive charts give you insights into your state of mind, how it has changed over time, and what factors may have influence such as exercise, sleep, and mindful minutes

Mental health assessments help you understand your current risk for depression and anxiety and if you might benefit from getting support

Screen Distance leverages the TrueDepth camera that powers Face ID to encourage you to increase the distance you view your device to reduce digital eye strain and can help reduce the risk of myopia in children

Sensitive Content Warnings can be enabled to prevent users from unexpectedly being shown images containing nudity in Messages, AirDrop, Contact Posters in the Phone app, and FaceTime messages

Expanded Communication Safety protections for children now detect videos containing nudity in addition to photos that children may receive or attempt to send in Messages, AirDrop, Contact Posters in the Phone app, FaceTime messages, and the system Photo picker

Improved sharing permissions give you even more control over what you share with apps, with an embedded photo picker and an add-only Calendar permission

Link tracking protection removes extra information from links shared in Messages, Mail, and Safari Private Browsing that some websites use in their URLs to track you across other websites, and links still work as expected

Accessibility

Assistive Access distills apps and experiences to their essential features in Phone and FaceTime, Messages, Camera, Photos, and Music, including large text, visual alternatives, and focused choices to lighten cognitive load

Live Speech lets you type what you want to say and have it be spoken out loud in phone calls, FaceTime calls, and for in-person conversations

Personal Voice enables users who are at risk of losing their voice to privately and securely create a voice that sounds like them on iPhone, and use it with Live Speech in phone and FaceTime calls

Point and Speak in Magnifier Detection Mode uses iPhone to read text out loud on physical objects with small text labels, such as keypads on doors and buttons on appliances

This release also includes other features and improvements:

Roadside Assistance via satellite lets you contact AAA to help you with vehicle issues when out of Wi-Fi or cellular range (iPhone 14, iPhone 14 Plus, iPhone 14 Pro, iPhone 14 Pro Max)

Pets in the People album in Photos surfaces individual pets in the album just like friends or family members

Photos Album widget lets you select a specific album from the Photos app to appear in the widget

Item sharing in Find My allows you to share an AirTag or Find My network accessory with up to five other people

Activity History in Home displays a recent history of events for door locks, garage doors, security systems, and contact sensors

Grid Forecast in Home shows when your electrical grid has cleaner energy sources available (Contiguous US only)

Grocery Lists in Reminders automatically group related items into sections as you add them

Inline PDFs and document scans in Notes are presented full-width, making them easy to view and mark them up

New Memoji stickers in Keyboard include Halo, Smirk, and Peekaboo

App Shortcuts in Spotlight Top Hit offer you app shortcuts to your next action when you search for an app

Redesigned Sharing tab in Fitness provides highlights of your friends’ activity like workout streaks and awards

Email or phone number sign-in lets you sign into your iPhone with any email address or phone number listed in your Apple ID account

New drawing tools in Freeform include a fountain pen, watercolor brush, ruler and more to create expressive boards

Crash Detection optimizations (iPhone 14, iPhone 14 Plus, iPhone 14 Pro, iPhone 14 Pro Max)

Some features may not be available for all regions or on all Apple devices. For more information, please visit this website:

https://www.apple.com/ios/ios-17

Some features may not be available for all regions or on all iPhone models. For information on the security content of Apple software updates, please visit this website: