Anthropic says it can read Claude's 'thoughts,' as detailed in new research paper — models observed to have a global workspace, revealing more of what makes LLMs tick

Anthropic's latest research reveals that its AI model, Claude, has an internal "J-space" resembling a global workspace akin to human cognitive processing. This discovery could lead to better oversight and more reliable AI, as it provides insights into how large language models operate. Understanding these mechanisms is crucial for developing more ethical and effective AI systems, potentially paving the way for improvements in AI transparency and accountability.