Practical wisdom

For each situation, these systems produce an explicit account of which values bear on it and which way each one points. They do this instead of collapsing a trade-off into a single number before the question is asked. Kaleido stops at this list, naming the values, rights, and duties involved and whether each supports or opposes the action. Chain-of-thought training on value conflicts carries the same material through to a verdict. The automated parliament splits the work across several delegates, each of whom speaks for a moral perspective. The delegates rewrite one another's answers and then agree on one together. Because the considerations are written out rather than hidden in a score, they can be analyzed at scale. For example, sorting hundreds of reasoning traces on trolley problems by the kind of reason given shows that models reason from duties in the chain of thought but explain themselves by consequences afterward.

Rather than committing to one ethical theory, the system holds several and settles their disagreements with a rule written down in advance. In one design, confidence is split across the theories, and these weights shape what a learning agent is rewarded for. In the other, the theories are treated as parties bargaining over lotteries, and the system takes the bargaining solution. Only the theories' verdicts on the case are inputs, and the rule that combines them does not change from case to case. This is deliberately the opposite of open-ended judgment. Its purpose is that the answer is predictable and can be debated before any particular case arises.

These papers argue that what is missing is not a procedure added at answer time but a trait that the model lacks, and that training should target that trait directly. One paper holds that fine-tuning instills surface behavior (refuse this, comply with that) but never the ability to reason when those instructions conflict. This is exactly the weakness that adversarial prompts exploit. The other paper holds that the harder gap is knowing which strategy to use for a problem that has no settled method. It outlines how that kind of judgment might be benchmarked, trained, and built. The two papers name different traits, and neither has been built. Both, however, aim to change what training instills rather than the form of any one answer.