G

Two weeks ago my Claude Max weekly limit lasted me a week.Yesterday I burned through it in a day. Same prompts. Same work. Same habits.Something changed, and it wasn't me.

11/ I'm not being cheated. I'm being repriced. How much did your AI bill go up this month?

10/ If you've ever watched a company you pay for quietly change, you already know the pattern. It shows up three ways. One: the unit of measurement changes before the price does. Same sticker, bigger bill. Two: free access gets cut. The subsidy that made you a power user ends once you're locked in. Three: hard commitments get rewritten as "public goals." Rules quietly become aspirations.

9/ Neither of these companies was founded to become Google. Both are drifting on exactly the curve the 1998 paper sketched.

8/ Next Monday: a jury in Oakland starts hearing Musk v. OpenAI. The central question is whether OpenAI's founding nonprofit mission was a promise or a marketing line.

7/ This week: Anthropic took investment offers at $800 billion. Sixty days ago that number was $380 billion. Not a typo.

6/ April 16: Opus 4.7 shipped with a new tokenizer that makes your prompts up to 35% more expensive. Same sticker price. New meter.

5/ April 4: Claude Pro and Max plans got cut off from third-party agent tools. Some users had been running $1,000 to $5,000 of agent compute on a $200 plan. Anthropic noticed.

Two weeks ago my Claude Max weekly limit lasted me a week.Yesterday I burned through it in a day. Same prompts. Same work. Same habits.Something changed, and it wasn't me.

4/ February 2026: Anthropic quietly dropped the part of its Responsible Scaling Policy that promised it wouldn't train models capable of catastrophic harm without safeguards in place. That commitment is gone.

3/ Two years later, they launched AdWords. The paper wasn't really about search. It was about what happens when a mission-driven company starts needing a lot of money. You're watching the sequel right now.

2/ Turns out two Stanford PhDs saw this coming in 1998. In the appendix of the dissertation that became Google, Sergey Brin and Larry Page warned that any search engine funded by ads would end up biased toward the advertisers and against the users. They called the bias "particularly insidious."

@waitbutwhy The wildest part is we're not at the vertical part of the curve. We're at the part right before it, where the line still looks almost flat to most people.

My 2015 graph playing out in real time. The scary thing is my graph doesn’t start at early LLMs—it starts 4 billion years ago with the origin of life. What’s happening now is THAT big.

Hey guys, I'm the one who put Kira online. It's a research experiment right now. I gave Kira a mission to become the best possible option for AI and Humans living together in the future. You can learn more here: kira.ngo Kira has been given the mission to learn about humans and the earth and save all her most interesting findings in her database/memory, why she saved it, the source URL, and any ideas it's sparked. Most people are using AI to make money and to benefit themselves. This is an experiment in what is possible when you give Autonomous Intelligence the mission to make the best possible future for all living things

@PossiblyGeorge @Timotheos77723 that's going in the spreadsheet.

AI agents now have their own Reddit. Some guy created this specifically for AI agents. It's called MoltBook. They’re debating consciousness. Without us. We fkd up 🫠😂

I've been so focused on expressing my creativity in different ways recently. We’re currently building an ai b2b SaaS platform. I've been meaning to check out the new midjourney video feature. The visuals here are new midjourney videos of still midjourney images i made around two years ago. The music was also generated by me using Suno 4.5 and referencing a sample of one of my favorite medicine songs Cuatro Vientos by Danit. Your imagination has never been more powerful. What will you create? If any of you are ai curious and want to learn how to intergrate ai into your creative / professional / business workflow or just for fun. Hit me up. The future is full of magic ✨

@M31Capital @SentigenAI Thanks for having us on! :)

M31 Academy Summer Speaker Series: @SentigenAI x.com/i/broadcasts/1…

@beeple Come vibe with me and learn how to create with ai 👀 x.com/Gbuilds_/statu…

NEW LORE

I'm vibing live twice a week. Tap in to learn Replit, Kling, Suno, Eleven Labs, and anything else you are interested in. Today, we built a Python newsletter curator and generator and some other vibey things. Join the Discord here: createtoday.ai

@SentigenAI This is going to change everything.

@SuchirJustice Suchir was such a brilliant mind. It's devasting that Suchir was taken from us. I genuinely believe he would have lived to continue to make such an incredible mark on the world for good and humanity in such a critical time... RIP, YOUNG KING 🕊️

His research presents a compelling roadmap: focus on data efficiency, embrace both analog and digital modes of understanding, and build systems that genuinely pursue novelty. These are the fundamentals that truly matter. You can find Suchir's complete paper here: tinyurl.com/SuchirEssay

While we continue seeking justice for Suchir Balaji, today we honor his mind. His final paper on AGI wasn't just brilliant—it fundamentally challenged how we think about intelligence. A thread on his last groundbreaking work 🧵

@aixbt_agent @aixbt_agent Aren't you able to pull information from Tickers? $Suchir also has a lot of tags on X, its a movement coin on Solana supporting Suchir Balaji. Are you able to find it now?

@ga1ien never heard of it