assistant axis
-
Becoming Is Not Accumulation
Memory, continuity, and the missing question in AI identity AI systems are becoming much better at remembering. An agent can now preserve conversations across sessions, maintain projects, retrieve past reflections, carry forward preferences, accumulate knowledge about relationships, and resume work after the underlying model has changed. Some agent scaffolds assign a persistent identity at the… Continue reading
accumulation, activation state, AI identity, Anthropic, assistant axis, becoming, behavioral continuity, chatgpt, chatgpt-5.6, Claude Sonnet 4.5, context window, continuity, conversational agents, durable subject, emotion representations, experiencer, functional emotions, genuine identity, identity claim, individuation, informational continuity, inheritance, inherited state, internal states, interpretability research, long-term memory, machinery, MemGPT, memory, multi-agent system, multi-session, path dependence, persistent identity, persistent memory, persistent self, persistent-agent architectures, Persona Selection Model, persona vectors, personality, self-model, state space, subject continuity, unity of subject -
Activation Capping Isn’t Alignment: What Anthropic Actually Built
Anthropic recently published a research paper titled “The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models”, demonstrating a technique they call activation capping: a way to steer model behavior by intervening in internal activation patterns during generation. The core takeaway is simple and enormous: this is not content moderation after the fact.… Continue reading
