I used GPT-6 Astra to break an unsolved cipher to one of Napoleon's generals that had gone unread for 217 years.
What makes this impressive isn't actually the codebreaking, but that Astra completed the entire multi-modal workflow in ~6 hours from a single image and goal.
1/6🧵
Anthropic CEO worries we're 6-12 months away from a swarm of AI agents capable of taking over the ENTIRE internet
Calls for immediate, global 1-2 year slowdown in AI development
Time to trust the people building the technology over the people merely investing in it
@TylerAlterman i enjoyed the dlc slightly less but i do feel you’d be doing yourself a disservice by not playing it. i found the exploration and mystery of the story more gripping in some ways than the main plot
So to recap:
- Israel knew about Oct 7
- Israel lets it happen
- Israel Apache helicopters maximize destruction & death
- Israel lies about war crimes and mass rape to manufacture consent for a genocide
- Israel buries all evidence
- Israel launches an ethnic cleansing war
- Israel cuts off water, food and supplies
- Israel blocks all foreign media & international oversight
- Israel targets media and medical staff and infrastructure
- Israel launches worldwide silencing propaganda campaign
- Israel expands land grab and illegal invasion into West Bank and Lebanon
- Israel pulls U.S. into illegal invasion of Iran
- Israel confirms plans for Gaza final solution
Yes, this result cost millions of dollars.
But remember that when @OpenAI announced o3 it cost ~$500,000 to score 87.5% on ARC-AGI 1. Today, Astra scores higher for ~$20.
In 2025 it took us and GDM an enormous amount of compute to achieve IMO gold. For the 2026 IMO, anyone with a $20/month ChatGPT subscription could do it.
Massively scaling test-time compute gives us a glimpse of the future. I believe that a year from now everyone will have an AI at their fingertips capable of solving problems of this caliber.
too buddhist to really be a rationalist
too christian to really be a buddhist
too new age to really be a christian
and, tragically, too rationalist to really be new age
The danger of selling to big companies, if you're a startup, is that they don't say no outright. They have months of meetings with you first. Since you hate meetings, that seems to you a sign of commitment. But it's not. They love having meetings! It's almost all they do.
Hats off to doctoral student and podcaster Zhong Shu in Beijing, China, who did this fascinating and smart interview with Christopher Nolan about THE ODYSSEY yesterday, originally posted to Bilibili.
A great listen/watch.
When a woman says she wants her man to be stoic and never need to be babied, she's rejecting the child in him. She's saying, "I need him to be my father."
When a man says he wants a woman he can fall apart in front of, who never needs his steadiness, he's rejecting the child in her. He's saying, "I need her to be my mother."
Not that it works in reverse: If you want a codependent hell of a relationship, indulge each other's inner child without boundaries, and swallow your resentment until you explode. If you want an independent hell of a relationship, completely reject each other's immaturity and weakness. And prepare to get dumped and lose all self-respect when the day inevitably comes that you lose your mind and throw a tantrum.
Or, you can take the Middle Way: take turns being mom and dad for each other when the other needs it. And work hard on yourself so you don't need it too often, and burn your beloved out.
Choose wisely!
[opens the portal to the godlike superintelligence that solves 87-year-old math problems and carries out autonomous cyberattacks]
“how long peanut butter good in fridge”
I’ve found that the Anna Karenina principle also applies to lying: all truth tellers feel the same freedom, all liars spiral into a bespoke hell because their twisted reality backfires in the exact way they tried to cheat it in the first place