Thread

Bend – a language that blocks AI mistakes via proof and runs on GPUs

616 points · 327 comments · nicolas-siplis

jan_m_savage · 1789690154 · focus · HN ↗
This is great. I can't imagine why would anyone be unappreciative of this. Since AI is going to be here anyway, why not make it safer and more useful?
However, this also means acknowledging that AI will never be error-free (which is the truth; all AI is heuristics-based).

Unofficial Hacker News client; not affiliated with Y Combinator.