AI Alignment is the trust problem between humans and AI. It is the most important techno-social challenge of our time. By building trust infrastructure as a public good, we aim to scale trust and safety at the pace of AI progress. This is just the beginning

i won’t let you down, humanity

The most fun image & video creation tool in the world is here. Try it for free in the Grok App.

great startup idea that i won't build myself but would love to invest in:. vercel for enclaves. start with fast deployments and evolve into building your own hardware to allow inference of large models inside. everybody needs this right now to guarantee privacy to users.

RT @carryhq_: The O-1 visa is for people with “extraordinary ability” but what does it actually take to get one?. Hear how @0xSigil, @kiril….

one of my profound core values is *endlessly decreasing not freedom*. freedom of movement, freedom of thought, freedom of creation and information. “not freedom” is important wording, since “freedom” is a subjective perception — you can be in prison and still feel free.

moved all my 2fa from phone to 1password and ordered yubikeys for 1password itself. be protected, it’s not hard.

source 1.

some new thoughts on privacy

this place has lost its way without me. rolling back in.

writing the whitepaper as a piece of art.

my warning from early december is happening?

it's going to be so great you can't even imagine.

is the only human network. don't trust other copycats.

i think people mean “control the outputs of the model” when they say that,. but ai alignment is ultimately a broader term—it’s about humans and ai coexisting in the most efficient way, supporting decentralization and opposing censorship. humans should collectively decide what the.

any new foundation built on specific principles will never be free of those principles, and everything built on top will inherit them.

i feel extremely grateful for everything i have in my life and the ability to make my life even better every day. not everything you do goes according to plan, but the love for the craft, the process, and the people around you magically leads to the best outcome after all.

ai ppl spend much less time thinking about ai agent registries than crypto ppl do, but they have better access and expertise to deploy them in the world.

whoever thinks that their path to user adoption is burning vc money in the form of grants to “build” on their protocol is a clueless makaka.

thx @byjmitch & @HF0Residency for the video.

young talented kids from sanctioned and low-income countries should have the right to great education and participation in the global economy. i was lucky (even though i dropped out of middle school), but millions aren’t. talent is universal. we should fix this.