ZDNet AI→ original

Anthropic releases Opus 4.8: model shows equal alignment with Claude Mythos

Anthropic released two new models — Opus 4.8 and Claude Mythos Preview. Testing showed both versions have the same level of misalignment — a safety metric measu

Anthropic releases Opus 4.8: model shows equal alignment with Claude Mythos
Source: ZDNet AI. Collage: Hamidun News.
◐ Listen to article

Anthropic released two new models — Opus 4.8 and Claude Mythos Preview. Testing showed both versions have the same level of misalignment — a safety metric measuring how well a model follows instructions. This confirms the consistency of Anthropic's approach to alignment and safety. For developers, it means both models are equally reliable and predictable in use.

ZK
Hamidun News
AI news without noise. Daily editorial selection from 400+ sources. A product by Zhemal Khamidun, Head of AI at Alpina Digital.
What do you think?
Loading comments…