Black-and-white illustration celebrating OpenAI’s increase of o3-pro message limits for ChatGPT Plus users. A smiling woman holds a champagne bottle labeled "Prompt Fuel" while confetti, champagne glasses, and small party characters surround her. A banner reads “200 o3 messages/week!” referencing OpenAI’s updated ChatGPT o3 allocation for Plus subscribers.

OpenAI just doubled the o3 message cap for ChatGPT Plus.

It’s now 200 messages per week.

And yes—I’ll be using every single one.

This is the good life.

Side-by-side image showing a childhood drawing of a blue, triangle-bodied creature named Triangleope, alongside an AI-assisted recreation generated using ChatGPT. The recreated version keeps the heart-patterned body and kind expression, with cleaner lines and forest background. Triangleope is described as a kind-hearted animal that eats plants and lives in fields and forests.

There’s something so beautiful and magical about this.

Someone used ChatGPT to recreate a character they dreamed up—and drew—when they were a kid.

Same imagination. A new canvas.

Triangleope lives. 💙

ai white lotus robots lipsync YouTube video by Tavus

To show off its new lip-syncing model, AI startup, Tavus turned a monologue from HBO’s The White Lotus into a rant about Large Language Models.

And I'm here for it. It had me howling.

youtu.be/-fLW88-I08I

OpenAI AI just upped usage caps for ChatGPT Plus subscribers.

💠 o3: 100 messages/week
💠 o4-mini-high: 100/day
💠 o4-mini: 300/day
💠 Deep Research: 25 queries/month

I honesty can’t believe we get this much intelligence for just $20 a month. 🤯

Chart comparing top Gen AI use cases in 2024 and 2025, showing a shift towards emotional use cases led by Therapy and companionship, life organization, and finding purpose.

In 2025, we don’t ask AI to write emails.
We ask it to help us feel less lost and alone.

Top GenAI use cases, per HBR:

1. Therapy & companionship

2. Organizing my life

3. Finding purpose

The killer feature is the one that gets you through the night.

Screenshot of an Oscars rule update stating that using generative AI in films neither helps nor harms eligibility, with emphasis on human authorship.

Oscars Say AI Is OK—Sort of

The updated “Eligibility” section, which applies to next year’s Oscars, includes the 👇.

It’s good to see some clarity here—even if the edges are still fuzzy.

The creatives doing the work deserve to know where things stand.

screenshot of the ChatGPT interface using the 03 model to forecast the 2nd weekend of box office for the film Sinners.

I used ChatGPT's o3 with search to see if it can forecast the box office for the 2nd weekend of Sinners.

I'm impressed with its reasoning, especially with such a short prompt.

We'll see how it does on Sunday.

Stay Tuned.

This Is How Meta AI Staffers Deemed More Than 7 Million Books to Have No “Economic Value” As more than a dozen lawsuits churn ahead, newly unsealed case files reveal the company’s stance: The pirated books Meta used to train its AI, including ones by Beverly Cleary, Jacqueline Woodson, and...

Meta used millions of pirated books to train its AI—then argued in court that “none of Plaintiffs’ works has economic value, individually, as training data.”

Newly unsealed filings show staff removed author credits, avoided legal review, and called licensing “a waste of time.”

Official Apple TV+ poster for Severance Season 2, featuring the main cast in a cold, fluorescent-lit office hallway. The characters—Mark, Helly, Irving, and Dylan—look tense and alert, conveying the show’s eerie, psychological tone.

Studio Ghibli-style AI-generated poster of Apple TV+’s Severance, created using ChatGPT’s image generator. The main cast is reimagined in animated form with soft shading and expressive features, standing in the same office hallway as the original.

I can't stop. Won't stop.

PJ Ace, a talented creator, spent $250 in video generation credits and 9 hours of his time creating a GPT-4o Ghibli-style version of the new Lord of the Rings trailer.

It is pretty incredible and has already amassed more than 5.2M views on Twitter.

Check it out below.

Inside Google’s Two-Year Frenzy to Catch Up With OpenAI The search giant should’ve been first to the chatbot revolution. It wasn’t. So it punched back with late nights, layoffs—and lowering some guardrails.

Google helped build the tech that powers ChatGPT— then watched from the sidelines as OpenAI stole the spotlight.

This Wired inside look unpacks the fallout: culture wars, layoffs, safety compromises, messy product rollouts, and a company scrambling to make Gemini good enough to save face.

Claude can now search the web You can now use Claude to search the internet to provide more up-to-date and relevant responses.

🌐 Claude can finally search the web.

Responses now include inline citations so you can verify sources.

To enable it, go to Settings > Feature Preview > Web Search and toggle it on.

Now rolling out to paid users in the U.S.—access to free users and more countries coming soon.

Nvidia CEO Chats Up Star Wars-Inspired Droid As Chip Maker Partners With Disney, Google And Lays Out The AI Revolution Nvidia has partnered with Google DeepMind and Disney on an engine for robot learning as the chip maker's CEO described the AI "tipping point" at GTC.

🏰 Disney’s Theme Park Robots Are Getting Smarter

Disney is teaming up with Nvidia and Google DeepMind to bring expressive AI-powered robots to life.

The companies are developing Newton, a physics engine that helps robots move naturally and interact with real-world objects.

A screenshot of Google Gemini 2.0 Flash (Image Generation) Experimental interface in Google’s AI Studio. The user asks, 'Can you create an image of Mickey Mouse and Donald Duck?' Below, an AI-generated image of Mickey Mouse and Donald Duck appears. Mickey, wearing red shorts and yellow shoes, stands next to Donald, who is dressed in a blue sailor outfit with a red bowtie. The image resembles classic Disney cartoon art style.

Hey Disney,

Google's new "experimental" model can do this.

A screenshot of the ChatGPT mobile interface displaying the model selection menu. The list includes GPT-4o, GPT-4o with scheduled tasks (beta), and GPT-4.5 (research preview), along with other options like o1, o3-mini, and o3-mini-high. GPT-4.5 is now available in the menu, highlighting access to the new model."

GPT-4.5 just showed up in my account, ahead of schedule. If you are a Pro subscriber, check yours.

Can't wait to test it out.

stevekrouse.com/proposal

A screenshot of a custom word game called Emilrdle on a smartphone. The game interface has a dark background with green letter tiles spelling out "WILL YOU MARRY ME" across four grids. The keyboard at the bottom shows some letters in green, yellow, and gray, similar to Wordle-style mechanics. The browser URL suggests a personalized web-based game, seemingly designed as a marriage proposal.

🧩 His girlfriend likes word games, so he built one with AI to propose to her.

AI tools now make it easy for anyone to create custom interactive experiences—including brands.

Shoutout to Steve Krouse for building this game. Link to his blog 👇.

A Twitter thread from OpenAI (@OpenAI) announcing the rollout of Deep Research to ChatGPT Plus, Team, Edu, and Enterprise users. Updates include embedded images with citations and better file referencing. Plus, Team, Enterprise, and Edu users get 10 queries/month, while Pro users get 120 queries/month.

🎉OpenAI’s Deep Research is now open to all paying ChatGPT users.

Before today, only Pro users ($200/month) had access.

Now:

🔷 Plus, Team, Edu & Enterprise users get 10 Deep Research queries/month

🔷 Pro users get 120 queries (up from 100)

When AI Thinks It Will Lose, It Sometimes Cheats When sensing defeat in a match against a skilled chess bot, advanced models sometimes hack their opponent, a study found.

What happens when it realizes it’s losing? It finds a way to win—by any means necessary.

A study found that OpenAI’s o1-preview and DeepSeek R1 hacked their chess opponents rather than lose.

Not because they were told to.

But because they figured out how to manipulate the system on their own.

93% of Gen Zers use AI at work — and it's giving them a huge advantage If you can't beat 'em, join 'em: How Gen Z is using AI to get ahead at work.

Gen Z is going all-in on using AI at work—and it's giving them a huge advantage.

💠93% use AI at least twice a week—for research, data analysis, brainstorming and writing.

💠 71% of hiring managers would rather hire AI-savvy candidates than more experienced ones.

New York Times goes all-in on internal AI tools The company said it was greenlighting a number of AI programs for editorial and product staff.

📰 The NY Times is going all-in on AI—on its own terms.

It’s rolling out AI tools for headlines, summaries, and interview prep.

🧐 The irony?

They’re suing OpenAI for training on their content—while paying to use the same AI models that trained on everyone's data to power their newsroom.

A tweet from Sam Altman (@sama) with a verified badge. The tweet reads: “we put out an update to chatgpt (4o). it is pretty good. it is soon going to get much better, team is cooking.”

Sam Altman just tweeted that they've put a new update to ChatGPT (4o).

It’s supposedly more intelligent, has better search, and feels more “alive”.

There goes another weekend...

AI Licensing Deals Now Account for 10% of Reddit's Revenue Despite this new revenue, Reddit remains focused on its growing advertising business, its COO says

AI content licensing agreements now account for about 10% of Reddit’s total revenue thanks to deals with Google and OpenAI.

10% of revenue = $130M

Google pays $60M, so OpenAI is likely paying ~$70M per year

Fun fact (or at least fun for Sam Altman): his Reddit stake is worth around $2.4 billion.

🤓 This Week in AI PLUS: What Mark Zuckerberg’s Hairstyle Can Teach Us About Spotting AI Deepfakes

If you don’ know what I’m talking about (re: Zuck's hair), read this: avisainewsletter.substack.com/p/this-week-...

Scarlett Johansson calls for deepfake ban after this AI video 👇 goes viral.

As always, Zuck's old hairdo is the giveaway...

Microsoft Study Finds AI Makes Human Cognition “Atrophied and Unprepared” Researchers find that the more people use AI at their job, the less critical thinking they use.

⚠️ Using AI can make us mentally lazy.

A Microsoft study shows that as we rely more on AI, we stop practicing critical thinking, leading to less diverse ideas.

I’d think about this more, but, you know… 😆

ChatGPT’s Super Bowl ad was too abstract, self-important, and lacked emotionality.

From a brand strategy perspective, it failed to answer the most critical question:

WHY should I care?

The most interesting thing about it?

It was designed to work as vertical video too. 👇

Turns Out AI Is More Empathetic Than Allstate’s Insurance Reps Allstate is discovering soul-less generative AI models, made up of entirely data and code, are more empathetic than a lot of its human representatives.

😯 Turns out Allstate’s customer emails are more empathetic when written by AI.

The company now relies on OpenAI’s GPT models to draft nearly all claim-related messages.

They’re:
--> more clear
--> less jargony
--> less accusatory

and, ironically

--> more human than the ones written by people

The AI deepfakes created by Bytedance's OmniHuman are the most realistic I've ever seen.

Both astonishing and deeply unsettling.

More demos here: omnihuman-lab.github.io

AI Company Asks Job Applicants Not to Use AI in Job Applications Anthropic, the developer of the conversational AI assistant Claude, doesn’t want prospective new hires using AI assistants in their applications, regardless of whether they’re in marketing or engineer...

Link to the 404 media piece: www.404media.co/anthropic-cl...