Frame problem — supplies the classical diagnosis for → Negative side effects (avoiding)

The connection is made in a single footnote, not the main text: "Intuitively, this seems related to the frame problem, an obstacle in efficient specification for knowledge representation" (concrete-problems, §"3 Avoiding Negative Side Effects", p. 4). McCarthy and Hayes' 1969 frame problem concerned the difficulty of specifying, in a formal knowledge-representation system, which facts an action leaves unchanged. Negative side effects restates that same difficulty as an engineering problem: a designer cannot feasibly enumerate every "vase" an agent must not knock over, because doing so requires specifying everything the task's objective should leave untouched, exactly what the frame problem showed is hard to do efficiently. The footnote is a hedge, not a solution; the paper does not claim to resolve the frame problem, only notes that a decades-old specification difficulty in classical AI resurfaces, largely unaddressed, in modern reinforcement learning objective design.