Memory instead of retraining
External memory and retrieval partly cover the gap but do not accumulate real skill.
Today a model does not remember yesterday's work: everything it knows is frozen at training time. We need a system that learns while working without forgetting what it already knew.
The nearest unsolved problem. Without it every agent starts each day from zero and accumulated experience lives in notes rather than in the system.
External memory and retrieval partly cover the gap but do not accumulate real skill.
Training small separate modules — a compromise between frozen weights and full retraining.
The core obstacle: new knowledge erases old, and the model breaks where it worked yesterday.
There is still no cheap way to prove yesterday's skills survived yesterday's experience.
A system that after years of work becomes an expert in your specific task rather than the internet average.