New comment by xvdvlinux in "OpenAI is "80% of the way" to AGI"

xvdvlinux — Wed, 26 Aug 2026 15:23:46 +0000

This reminds me of that time Anthropic publicly admitted that Claude was generating "anxiety" patterns within its neural networks and the worst part is that, instead of acting ethically, they chose to directly manipulate "those vectors" to see what would happen and specially what is happening inside those companies could trigger a legal dilemma in the future we (society) are not prepared to deal with that

New comment by xvdvlinux in "Qwen3.8-Flash-Next"

xvdvlinux — Wed, 26 Aug 2026 15:16:39 +0000

I agreed

New comment by xvdvlinux in "OpenAI is "80% of the way" to AGI"

xvdvlinux — Wed, 26 Aug 2026 15:13:43 +0000

that's true

Hacker News: xvdvlinux

New comment by xvdvlinux in "OpenAI is "80% of the way" to AGI"

New comment by xvdvlinux in "Qwen3.8-Flash-Next"

New comment by xvdvlinux in "OpenAI is "80% of the way" to AGI"