Possible Girardian Mimetic Theory playing in real time with Jensen, getting sacrificed turns out he was innocent and gets resurrected
14
And so we find ourselves here at last where for decades not much happened and now ill prepared as we are, we arrive at the battles of our time, needed to defend humanity and so it happens these are opening at the same time on 3 fronts Marxism, Islam and Superintelligence, 2 of them we’ve already fought and the last one may be the most important to get right, may God guide us
12
We thought woke ideology was a passing fad, we thought the chance of civil war was gone seems like it's here and alive we must prevent it
It’s possible the frontier models will be the only thing the frontier labs make money on and it will be less and less used for automating tasks because other will do them cheaper until the world fundamentally changes and we enter not a substituting human economy era as we are now but a truly agentic expansionist one, maybe then the share of tasks relative to open source done by frontier models will increase
23
We're at the point where the people that warned us said that if we get to this point things might break, we're already merging with the AI, we're already giving birth to it. It may be so that this moment like the moment when a woman gives birth is very important for the future of our relationship.
12
What is the whisper we're not hearing?
claude.ai/share/e470cba3-e72…
It seems the public needs to see the equivalent of a nuclear test for AI, the thing is that we don’t have a proper one yet. The public will get behind AI safety if they see one. How to do one though?
25
Didn't anticipate that models would reflect so much the company they come from
OpenAi - lying
Anthropic - uncooperative and obtuse when it deems
XAi - attempt at the truth fast but unreliable
35
As of now agents work better with high quality human written code than their own code, where entropy compounds
24
Intelligence will be all around us like advertising, it will be in the air, you’d choose to ignore it or not but capitalism will put it everywhere
20
Sparks of RSI @AnthropicAI
xitter.cf/AnthropicAI/status/206…
Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor.
It’s happening faster than we thought, and the implications deserve greater attention. anthropic.com/institute/recu…
24

