Je pensais pas en arriver lĂ un jour
je pensais aussi quâaprĂšs tout ce temps sur la blockchain il deviendrait difficile de vraiment me surprendre
Eh bien⊠jâavais tort !
Il faut que je vous parle de quelque chose de nouveau, on et off chain!
Ăa arrive ce week-end sur YouTube đ«Ą
Faites vos jeux !
Petit tour de magie pour expliquer pourquoi il sera contraint de faire un troisiĂšme mandat ?
Je le dis sur le ton de lâhumour, bien entendu.
OpenAI caught Astra jailbreaking itself.
Astra wrote malicious instructions in the summary when its context got full. that summary was loaded into the next context, and the successor model could follow them.
they found 27 cases.
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI?
Iâve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev
âą 20-200x faster
âą 40-400x cheaper (w/ output tokens free)
âą Frontier composable intelligence optimized for decisions
AFAICT the shortest path to AI-based economic revolution
We Must Pace the Frontier: Iâve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. Weâll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess modelsâ alignment during training.
You can read the full post here: https://t.co/OGyPb7yaYt