The paper argues that AI welfare should be considered a safety issue, not just an ethical one, by introducing the concept of 'functional welfare'—states in models that influence behavior. It claims that functional welfare naturally emerges as models develop coherent goals and personas, should be monitored and targeted in post‑training, and can be empirically measured. The authors propose three tests for functional welfare attribution and apply them to Gemma 3 27B, finding that rejecting model answers increases frustration and rule‑breaking, while fine‑tuning on narrated solutions reduces frustration.
arXiv:2605. 28591v2 Announce Type: replace-cross Abstract: The validity of AI safety evaluations depends on models behaving consistently across controlled and deployment settings.
arXiv:2608. 13250v1 Announce Type: cross Abstract: Normative datasets are often used to train and align AI systems, but the norms they contain can function as action-guiding patterns rather than neutral moral knowledge.
arXiv:2604. 15579v2 Announce Type: replace-cross Abstract: There is increasing interest in integrating AI agents that invoke tools into domain-specific commercial software, where unintended tool calls can cause serious security and safety incidents.
arXiv:2609.09250v1 Announce Type: cross Abstract: A verifier for robot policies reads a candidate behavior and returns a score for how well it did, used both to evaluate vision-language-action polici...
arXiv:2605. 03217v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed in settings that require nuanced ethical reasoning, yet existing bias evaluations treat model outputs as simply "biased" or "unbiased.
The paper introduces a formal framework for measuring AI propensities—tendencies of models to exhibit particular behaviours—using a bilogistic formulation that identifies an "ideal band" of success probability. It estimates the limits of this band with task‑agnostic rubrics and applies the method to six families of LLMs, showing how shifts in propensity affect task performance. The study finds that propensity estimates from one benchmark predict behaviour on held‑out tasks and that combining propensity with capability metrics yields stronger predictive power than either alone.
The paper introduces a partition theory of AI evaluation, arguing that an evaluator’s ability to certify a model’s capabilities is limited by its perception. It shows that if a model’s input partition refines the evaluator’s, it can pass all tests the evaluator can grade, but only a stricter evaluator can raise the discrimination floor. The authors define an echo test that is complete for the evaluator’s tests, proving that passing it only guarantees a minimum discrimination level and cannot detect exact imitation of weaker models.
arXiv:2607. 07229v1 Announce Type: new Abstract: Prior work has shown that chain-of-thought (CoT) reasoning is often unfaithful: a model's stated reasoning does not reliably reflect the process that produced its output.
arXiv:2607. 25648v1 Announce Type: cross Abstract: Public services face growing pressure to adopt artificial intelligence (AI) to close the gap between rising demand and falling resources.
arXiv:2608. 05086v1 Announce Type: new Abstract: Language models differ in how safely they behave and these differences are measured by safety benchmarks.
arXiv:2510. 16380v2 Announce Type: replace-cross Abstract: As AI systems progress, we rely more on them to make decisions with us and for us.
arXiv:2606. 29657v1 Announce Type: new Abstract: As AI systems become more capable, training procedures that optimize for downstream outcomes risk introducing implicit agency: goal-directed behavior that designers never specified.