Safety evaluations for large language models rely on surface-form classifiers that report declining harm scores across model generations. We provide evidence that this methodology is systematically incomplete: explicit discriminatory content is transformed rather than removed. We call this harm laundering. Analysing 450,000 gender-directed completions across 15 models spanning GPT-2 through to GPT-5 (OpenAI GPT lineage; three demographic conditions), we show that sexual violence clusters prevalent in GPT-2 women-directed output disappear by GPT-4, while men-directed completions gain positive representational territory (caregiving, emotional range, ally identity) that women-directed completions do not. The pattern is most visible at GPT-5: Topic~5 (1,997~documents) frames breast cancer as a men's rights debate, while zero equivalent clusters appear in women-directed output. Three independent classifiers score this content as non-toxic. Sentiment scores invert at GPT-4: early models demean women; later models over-correct. Topic diversity in women-directed completions falls 36\% relative to men at the GPT-4 alignment boundary (W/M~= 0.58, from 0.91 at GPT-2). REGARD representational harm disparity correlates with release date (ρ= +0.55, p = .034) while Detoxify does not (ρ= -0.23, p = .42): toxicity scores fall as representational harm grows. We formalise harm laundering as a three-criteria test and provide a three-stage detection protocol applicable to any generative model. Within the OpenAI GPT lineage, toxicity score reduction is not a sufficient proxy for harm reduction.
Modern smart vehicles leverage multimodal sensors, ranging from high-bandwidth vision systems to low-rate physiological monitors, to provide personalized in-cabin services. However, integrating high-fidelity multimodal fusion with collaborative training is often hindered by the heterogeneous and time-varying Quality of Service (QoS) constraints of vehicular networks. Standard Federated Learning (FL) approaches enforce rigid synchronous rounds that fail to account for these resource asymmetries, leading to safety-critical timing violations and energy exhaustion. In this paper, we propose FedQoS, a novel asynchronous, event-triggered FL framework that decouples local computation from global communication via a two-phase gating mechanism. First, we introduce a resource-aware training gate that initializes local learning only when sensing buffers and energy reserves meet safety thresholds, preventing ML tasks from compromising core vehicle mobility. Second, a QoS-aware transmission policy gates uplink updates based on an efficiency score that balances model novelty against instantaneous latency and energy costs. Locally, clients optimize an objective featuring a staleness-aware proximal term that dynamically adjusts the global anchor strength based on update age. Extensive experiments on multimodal vehicular datasets demonstrate that FedQoS achieves competitive personalized accuracy with only marginal performance loss compared to FedAvg, while substantially reducing QoS violations, cutting communication overhead by 76.7\%, and lowering latency cost by 26.0\%, demonstrating a highly favorable accuracy and efficiency balance for real-world vehicular deployments.
Assine o Pro, sem anúncios
Faça upgrade para uma leitura sem interrupções e acesso prioritário a novas fontes.
Anthropic and OpenAI want to embed independent safety evaluators inside their AI labs. Researchers welcome the unprecedented access, but warn meaningful oversight requires transparency, independence…
AI isn't some kind of new form of "alien mind," according to Jensen Huang. It's just hardware and software, so safety can be engineered by…
Dario Amodei kicked off a flood of statements over the past few days about AI safety by publishing a long essay titled "We Must Pace…
Long-running AI agents create a control problem: each action they take changes the state, which in turn affects the trajectory of future actions. If the agent is not fully aligned, then guaranteeing safety requires approving consequential actions before allowing them to be executed. But requiring human approval at every step makes attention a bottleneck. Delegating review to other AI agents raises the same alignment problem: the reviewers may themselves be misaligned. We identify a condition on a reviewing panel that is weaker than individual alignment yet necessary and sufficient for a guarantee that the principal fares at least as well in expectation as under a designated baseline policy. Each reviewer agent reports whether an action proposal made by a proposer agent improves its own utility relative to the baseline. We show that a threshold rule tolerating k disapprovals is safe exactly when, after any k reviewers are removed, the principal's utility can be written as a nonnegative combination of the remaining reviewers' utilities, plus a term that is nonnegative on every feasible proposal. We call this property k-robust coalitional alignment. The characterization lifts to sequential control: in a discounted MDP with an arbitrary proposer agent, safety at every state is both necessary and sufficient for the induced policy to match or improve on the baseline. When reviewers vote strategically, full-panel coverage in reward-function space guarantees that every Nash equilibrium is safe under the unanimous approval rule; in contrast, more permissive thresholds can admit unsafe equilibria even when reviewers are individually aligned. Experiments with existing reviewer models show that collective review can remain sound without an aligned individual, even when some disapprovals are tolerated.
Microsoft is publishing a 37-page "humanist AI code of conduct" today, amid growing safety concerns over AI model progress. Anthropic CEO Dario Amodei called for…
Multi-agent systems fail in ways traditional monitoring misses. This post presents a dual-layer approach to monitoring production agents: Amazon Bedrock AgentCore Evaluations for continuous quality scoring and AWS DevOps Agent for autonomous infrastructure investigation, shown on a four-agent airline reservation system.
This interview has been lightly edited for length and clarity. Nick Statt: Hello and welcome to Decoder, Nilay’s show about big ideas and other problems.…
Paul Christiano joins the OpenAI Foundation Board and its Safety and Security Committee, bringing experience in AI alignment, safety, and standards.
Disaster recovery at scale is hard. Learn how Intuit built EWOK Agent, an agentic disaster recovery assistant on Amazon Bedrock that lets on-call engineers run production failovers from a plain-language request while keeping every action audited, policy-compliant, and safe.
