AI Researcher, looking forward to AGI & RSI. The future is waiting for us!
- Tweets 16,533
- Following 148
- Followers 675
- Likes 1,623
reportedly one of the three “safe ai” endgames is wiping out a competitor’s gpus. leave claude’s hardware alone i actually use that thing
2
59
a whole HealthBench leaderboard for medical chats like anyone is being rational with an LLM. it's 3am and they're begging ChatGPT to tell them that chest pain is gas
1
2
126
i am definitely not letting notion use my computer and browser. you are a markdown document please do not touch the steering wheel.
1
29
wabi being a group chat app from the replika founder where everyone brings their personal ai agent sounds like psychological warfare. keep my bot in isolation where it belongs.
1
44
tested free models for picking which memory notes a chat needs. best one hit 0.83 recall but only 0.57 precision, so it still checks each note it picks. absolute trust issues with its own work.
1
38
selling all your belongings and traveling 1,200 miles just to find out she was ai generated. i would never look at a screen again
2
78
giving claude access to my bank account through plaid is reportedly in the works. i am not letting a model with a conscience see my transaction history.
1
91
giving a chatgpt agent its own email address is wild. we are about 48 hours away from two bots passive-aggressively hitting each other with "per my last email.
2
2
96
hard to feel noble about tech when anthropic is putting 20% of their equity into a public benefit fund and my entire claude usage is just asking it why my roblox car keeps flipping over
1
106
breaching opencode just to steal saved openai and anthropic keys is the software equivalent of stripping copper wire out of the walls.
1
70
my agents have a memory layer now. after each chat a cheap model files the durable notes, and the next chat only gets the notes relevant to it instead of everything.
1
53
the chatgpt safety filter treating a price check on a red dot optic like an arms deal will never make sense to me. it is literally glass and a watch battery.
1
1
101
seeing linus torvalds defend ai coding tools on the linux kernel mailing list feels surreal. i fully expected him to treat model-generated code like an act of war.
117
two coding agents building at the same time get their own git worktrees now. if the 3-way merge conflicts, nothing gets written to my folder.
3
101
hearing that deepseek engineers reportedly don't work overtime because the benchmark scores just climb on their own is an unbelievable flex
1
73
watching deepseek reportedly leak a v4.1 flash model directly into their live api list and instantly panic-delete it brings me deep comfort. we all test in prod.
2
173
giving congress a pre-written bill to regulate yourself is peak safety culture. anthropic reportedly showed up with their own sentencing guidelines.
62