context retrieval, reasoning models
Both of these are examples of more intelligent systems built on top of an LLM where the LLM itself has no state.
Thirdly, the AI can use the environment as state in ways which would be nearly impossible to fully trace or mitigate.
Again same problem—AI tool use is mediated by text input and output, and the validator just needs access to the LLM input and output.
Both of these are examples of more intelligent systems built on top of an LLM where the LLM itself has no state.
Again same problem—AI tool use is mediated by text input and output, and the validator just needs access to the LLM input and output.