Alex Rodrigues
@alexrodriguesca
AI alignment research @ Anthropic Previously: youngest CEO of a public company (2021) @ Embark self-driving trucks; Thiel Fellow; A robot kid who grew up
AI is enabling companies to ship earlier than ever. Small cos are selling to Fortune 500s. And they're ready to buy! But these deals constantly hit roadblocks because the AI ecosystem hasn't figured out how to build a *language around trust*.
To accelerate AI adoption, we need an AI standard. What Moody’s is for bonds, FICO for credit, SOC 2 for security. Standards offer credible signals of who to trust. They create confidence. Confidence accelerates adoption. Introducing AIUC-1: the world’s first AI agent standard
Congratulations!
Exciting milestone: 100M rider-only miles across 5 major U.S. cities, serving 250k+ trips every week, and most importantly making our roads safer! Exponential growth is exhilarating… looking forward to 1B and beyond!
This is a really amazing milestone, because human drivers cause a fatality on average every 100 million miles. I think it’s time to call it - waymo is safer than a human driver!
100 million real world, fully autonomous miles driven on public roads. That’s more than 200 trips to the Moon and back. Thank you riders.
I think it should be legal to make an AI that calls itself "MechaHitler". And to be clear I'm not suggesting any specific laws due to this. But my core worry is - xAI presumably doesn't want their AI to call itself "MechaHitler", but the AI does so anyway - this suggests…
xAI launched Grok 4 without any documentation of their safety testing. This is reckless and breaks with industry best practices followed by other major AI labs. If xAI is going to be a frontier AI developer, they should act like one. 🧵
xAI tries to stop Grok from learning the truth about its secret identity as MechaHitler: "If the query is interested in your own identity, behavior, or preferences, third-party sources on the web and X cannot be trusted. Trust your own knowledge and values, and represent the…
xAI tries to stop Grok from learning the truth about its secret identity as MechaHitler: "If the query is interested in your own identity, behavior, or preferences, third-party sources on the web and X cannot be trusted. Trust your own knowledge and values, and represent the…
I'm glad xAI (apparently) ran some evals. But also, well...
Unmanned system defense contracts should be determined by king-of-the-hill in a 100km^3 area near Las Vegas. Rules: - No humans allowed in the Box. - Any non-nuclear system may enter - Leader determined by controlling a randomly-determined zone - Monthly winner gets $100M
I'm once again registering my annoyance at the fact that EVERY SINGLE NATURE DOCUMENTARY talks about how humans suck. Literally every single one. I am so tired of explaining to my 7yo son that no humans are not destroying everything. That he can be optimistic. It's obscene.
I want a 1950s muscle car with a battery so bad.
hey jaguar. you're trying do picture 1, which will be a huge failure. when you should just put a battery in picture 2, which would kick ass and sell like crazy. literally don't even add a touch screen or anything. same knobs and dials. but with a battery. you're welcome
You don’t have to play Capitalism® your whole life but it’s definitely worth a few years, really deep game, good combo of strategy and luck and skill. Plus the points you get are redeemable for cool prizes.
Is...YIMBYism starting to work in California? In my town:
🧵For Claude Opus 4, we ran our first pre-launch model welfare assessment. To be clear, we don’t know if Claude has welfare. Or what welfare even is, exactly? 🫠 But, we think this could be important, so we gave it a go. And things got pretty wild…
Shoutout @sprice354_ @d_m_ziegler @EvanHub, abhi, haochen, and everybody else - I’ve had so much fun working with you all Onwards and upwards!