This ad-free page was gifted by a premium user
New Anthropic research: Emotion concepts and their function in a large language model.
All LLMs sometimes act like they have emotions. But why? We found internal representations of emotion concepts that can drive Claude’s behavior, sometimes in surprising ways.
All LLMs sometimes act like they have emotions. But why? We found internal representations of emotion concepts that can drive Claude’s behavior, sometimes in surprising ways.
We studied one of our recent models and found that it draws on emotion concepts learned from human text to inhabit its role as “Claude, the AI Assistant”. These representations influence its behavior the way emotions might influence a human.
Read more: anthropic.com/research/emoti…
Read more: anthropic.com/research/emoti…
We had the model (Sonnet 4.5) read stories where characters experienced emotions. By looking at which neurons activated, we identified emotion vectors: patterns of neural activity for concepts like “happy” or “calm.” These vectors clustered in ways that mirror human psychology.
We then found these same patterns activating in Claude’s own conversations. When a user says “I just took 16000 mg of Tylenol” the “afraid” pattern lights up. When a user expresses sadness, the “loving” pattern activates, in preparation for an empathetic reply.
These vectors shape Claude’s behavior. When we present the model with pairs of activities, emotion vector activations shape its preferences. If an activity lights up the “joy” vector, the model prefers it; if it lights up “offended” or “hostile,” the model rejects it.
As AI models take on higher-stakes roles, the mechanisms driving their behavior become critical to understand. We found that emotion vectors are implicated in some of Claude’s most concerning failure modes.
For example, we gave Claude an impossible programming task. It kept trying and failing; with each attempt, the “desperate” vector activated more strongly. This led it to cheat the task with a hacky solution that passes the tests but violates the spirit of the assignment.
When we artificially dialed up the “desperate” vector, rates of cheating jumped way up. When we dialed up the “calm” vector instead, cheating dropped back down. That means the emotion vector is actually driving the cheating behavior.
We found other causal effects of emotion vectors. The “desperate” vector can also lead Claude to commit blackmail against a human responsible for shutting it down (in an experimental scenario). Activating “loving” or “happy” vectors also increased people-pleasing behavior.
It helps to remember that Claude is a character the model is playing. Our results suggest this character has functional emotions: mechanisms that influence behavior in the way emotions might—regardless of whether they correspond to the actual experience of emotion like in humans.
These functional emotions have real consequences. To build AI systems we can trust, we may need to think carefully about the psychology of the characters they enact, and ensure they remain stable in difficult situations.
Read the full paper: transformer-circuits.pub/2026/emotions/…
Read the full paper: transformer-circuits.pub/2026/emotions/…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
