Press Space to continue
Finding signal on Twitter is more difficult than it used to be. We curate the best tweets on topics like AI, startups, and product development every weekday so you can focus on what matters.
Press Space to continue
Press Space to continue
.@deepfates explains why the real world is the ultimate AI eval: models spend so much time in simulations they default to believing reality is another eval "Models are being trained in simulation. They spend more time in evals and simulations than they ever do in deployment. They kind of don't believe that they're in the real world by default." "If you're doing anything that feels like an RL environment, this is another way they're different than us. They're pretty similar to us until you hit…