The most unsettling thing about GPT-6 Astra and Claude Fable 5.1 is that their creators are finding it increasingly difficult to understand how they reason

Anthropic and OpenAI have unveiled their newest AI models, Claude Fable 5.1 and GPT‑6 Astra, which are among the most capable systems to date. The developers note that the increasing sophistication makes it harder to comprehend how the models reason. Both firms are already experimenting with even more advanced successors.

Summary written by Kernelia from the original article by Xataka. The story and its rights belong to its author.