Rather than causing mental atrophy, AI can be a 'prosthesis for your attention.' It can actively combat the natural human tendency to forget by scheduling spaced repetitions, surfacing contradictions, and prompting retrieval. This enhances cognition instead of merely outsourcing it.

Karpathy identifies a key missing piece for continual learning in AI: an equivalent to sleep. Humans seem to use sleep to distill the day's experiences (their "context window") into the compressed weights of the brain. LLMs lack this distillation phase, forcing them to restart from a fixed state in every new session.

Retrieval-Augmented Generation (RAG) is just one component of agent memory. A robust system must also handle dynamic operations like updating information, consolidating knowledge, resolving conflicts, and strategically forgetting obsolete data.

Instead of treating memory as a component, adopt a "memory-first" approach when designing agent systems. This paradigm shift involves architecting the entire system around the core principles of how information is stored, recalled, and forgotten.

To manage context effectively, an AI OS can run a nightly routine ('dreaming') that reviews daily memory files, compresses key information, and saves it into a long-term memory file. This process mimics human memory consolidation, preventing context loss over time.

A new OpenClaw feature called "dreaming" allows the AI agent to process information and consolidate memories overnight while inactive. This concept, borrowed from human neuroscience, aims to improve the agent's long-term learning and performance without requiring active user input, mimicking how humans process experiences during sleep.

A key weakness of LLMs, the tendency to forget details in long conversations ("context rot"), is being overcome. Claude Opus 4.6 scored dramatically higher than its predecessor on this task, a crucial step for building reliable AI agents that can handle sustained, multi-step work.

Google's Titans architecture for LLMs mimics human memory by applying Claude Shannon's information theory. It scans vast data streams and identifies "surprise"—statistically unexpected or rare information relative to its training data. This novel data is then prioritized for long-term memory, preventing clutter from irrelevant information.

Long-running AI agent conversations degrade in quality as the context window fills. The best engineers combat this with "intentional compaction": they direct the agent to summarize its progress into a clean markdown file, then start a fresh session using that summary as the new, clean input. This is like rebooting the agent's short-term memory.