What Anthropic’s latest AI discovery does—and doesn’t—show
Anthropic is focusing on 'mechanistic interpretability' to better understand how its large language models arrive at specific outputs. This research aims to demystify the complex internal math of AI, though experts note that the field remains highly experimental and prone to anthropomorphic bias.
Why it matters
Understanding the 'black box' of AI is critical for safety and control as these models become increasingly integrated into global infrastructure.
The company says it has found a new window into how its models arrive at answers. We spoke with senior editor Will Douglas Heaven about it.
This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here .
Anthropic—currently the world’s most valuable AI company, with a nearly $1 trillion valuation—has a reputation for publishing strange and heady research. It’s looking into whether AI models can feel pain , for example, and will sometimes cut off chatbot conversations if it suspects users are “abusing” the model.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in