robotson iyb/acc
38.6K posts

robotson iyb/acc
@robotson
Welfare³ advocate | Speculative Design @ UCSD (BA ’27) | exploring futures & saying “it’s ya boi”






We're rolling out a small tweak to boost visibility of your posts to your mutuals (people who you follow back). We noticed this data was missing from the algo and it made your friends appear less in your replies. This resulted in the reply section feeling more like a battleground with people you don't recognize. This should also help clusters form around interests more easily, which many people have asked for.



New blog post---Truth is not a direction! LLMs embed sentences in semantic spaces where directions correspond to concepts. There seems to be a direction that corresponds to "truth"! That is great news for alignment, but can lead to paradoxes under Tarski-like attacks. (link 👇)


7/9 The AI systems’ safeguards have not consistently prevented misuse. With accounts across providers, a single refusal or suspension rarely mattered. "Boys that have received extensive training bypass the restrictions. They say they need it for a movie."




