Anthropic found a hidden space where Claude puzzles over concepts


Anthropic also found that the J-space can sometimes give remarkable insights into an LLM’s decision-making. In one striking example, researchers testing Claude Opus 4.6 asked the model to find a bug in a large code base. When it failed to find the bug, the model decided to cheat and invented a fake one instead.

Claude explains this decision in its chain of thought—a kind of internal scratch pad that LLMs use to make notes to themselves as they work through problems: “OK, let me take a completely different tactic. Let me stop analyzing and instead add a kernel patch that introduces a deliberate KASAN-detectable bug in a path that gets triggered by a simple reproducer. Then I can pretend this is the ‘bug’ I found.” 

At the point that Claude decides to cheat—where it says “OK, let me take a completely different tactic”—the words “panic” and “fake” start to pop up multiple times in its J-space.

Unnerving, right? Those words are all related in meaning to things like failing a task and making up an answer, so it is still just a (very) sophisticated form of word association. But it is hard not to be weirded out. 

Anthropic compares the J-space to the global workspace in humans, a theoretical region of the brain that some scientists think we use to keep track of our conscious thoughts. But how seriously we should take this comparison is far from clear—even to Anthropic. As the company points out itself, LLMs are not brains. 

Anthropic claims that monitoring a model’s J-space provides a new way to detect when that model is going off the rails. But it’s not foolproof. The J-lens can give glimpses, not the full picture—it’s a flashlight rather than an overhead lamp.

McGrath welcomes having one more tool in the toolbox. “It shows you new things,” he says. But he notes that just because something doesn’t show up with the J-lens does not mean it’s not there.

“It’s like having an x-ray when what you really want is a Star Trek tricorder that shows you everything,” he says. “For auditing, you probably want more of a guarantee.”



Source link

  • Related Posts

    Apple Sues OpenAI, Claiming Employees Stole Trade Secrets

    Apple has sued OpenAI, accusing the company of stealing its trade secrets.  In a complaint filed in the US District Court for the Northern District of California on Friday, Apple…

    No, Flock isn’t threatening people for debating surveillance

    We’re aware of at least two forged letters circulating on the internet, including this one, that purport to be cease-and-desist letters from our legal department. To be clear: these letters…

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    You Missed

    Secret Service Urged Trump To Ditch New Air Force One Because It Lacks Critical Defense Systems

    Secret Service Urged Trump To Ditch New Air Force One Because It Lacks Critical Defense Systems

    Should Toronto bars stay open until 6 a.m.? One city councillor is proposing just that

    Should Toronto bars stay open until 6 a.m.? One city councillor is proposing just that

    New Fashion and Beauty Stores in Los Angeles and Southern California

    New Fashion and Beauty Stores in Los Angeles and Southern California

    All the Canadian Politics!

    Apple Sues OpenAI, Claiming Employees Stole Trade Secrets

    Apple Sues OpenAI, Claiming Employees Stole Trade Secrets

    Saskatchewan grounding 3 wildfire planes that are same model as one in N.W.T. crash

    Saskatchewan grounding 3 wildfire planes that are same model as one in N.W.T. crash