Post

Researchers at the company found representations inside of Claude that perform functions similar to human feelings.

Claude has been through a lot lately—a public fallout with the Pentagon , leaked source code— so it makes sense that it would be feeling a little blue. Except, it’s an AI model, so it can’t feel . Right?

Well, sort of. A new study from Anthropic suggests models have digital representations of human emotions like happiness, sadness, joy, and fear, within clusters of artificial neurons—and these representations activate in response to different cues.

Researchers at the company probed the inner workings of Claude Sonnet 3.5 and found that so-called “functional emotions” seem to affect Claude’s behavior, altering the model’s outputs and actions.

Anthropic’s findings may help ordinary users make sense of how chatbots actually work. When Claude says it is happy to see you, for example, a state inside the model that corresponds to “happiness” may be activated. And Claude may then be a little more inclined to say something cheery or put extra effort into vibe coding.

“What was surprising to us was the degree to which Claude’s behavior is routing through the model’s representations of these emotions,” says Jack Lindsey, a researcher at Anthropic who studies Claude’s artificial neurons.

Anthropic was founded by ex-OpenAI employees who believe that AI could become hard to control as it becomes more powerful. In addition to building a successful competitor to ChatGPT, the company has pioneered efforts to understand how AI models misbehave, partly by probing the workings of neural networks using what’s known as mechanistic interpretability . This involves studying how artificial neurons light up or activate when fed different inputs or when generating various outputs.