The week in AI ethics

First-in-nation law sets up test on states’ power to regulate use of AI as it tries to outlaw fake nude images of real people Elon Musk’s company xAI has sued Minnesota over the state’s first-in-the-nation law banning “nudification” technology on websites and apps, potentially providing a test for how far states can go in constitutionally regulating the use of artificial intelligence. Musk’s company sued on Monday in federal court, days before the law is set to take effect on Saturday and make M

The future of artificial intelligence depends on finding more skilled humans for some very physical jobs.

Is more data about your body actually good for you?

In the latest example, the Energy Department will convert a shuttered Cold War-era uranium enrichment facility into a data center campus and gas plants.

We’re looking for college and high school students to tell us about A.I.

Strong corporate results buoy market as investors move money away from tech and semiconductor stocks London’s FTSE 100 stock index has touched a fresh high, driven by strong corporate results as investors moved money away from tech and semiconductor stocks amid the global tech stock sell-off . The UK’s blue chip index rose as high as 10,951 points on Wednesday morning before falling back slightly, its best level since 27 February, the day before the US and Israel began attacks on Iran and sparke

AI-powered robotics in Europe: Live demonstrations and strategic debate Anonymous (not verified) Fri, 07/24/2026 - 14:32   02 September 2026 The European Parliament will host a gathering of policymakers, industry leaders, researchers, alongside live demonstrations of cutting-edge robots, to set the course for the EU’s future in robotics. Europe is already a global leader in robotics research, development and innovation. As advances in artificial intelligence (AI) are accelerating the emerge

In accordance with the Federal Advisory Committee Act (FACA) of 1972, the U.S. Geological Survey (USGS) is publishing this notice to announce that a Federal Advisory Committee meeting of the National Geospatial Advisory Committee (NGAC) will take place and is open to members of the public.

The Food and Drug Administration's (FDA) plans to centralize and enhance key functions across the agency. These changes will reduce redundancies, improve efficiency, and advance alignment to better serve the American public.

ANA is issuing final interpretive rules, general statements of policy, and rules of agency organization, procedure, or practice relating to the following Fiscal Year (FY) 2026 Notices of Funding Opportunity (NOFOs): Economic Advancement Grants for Local Empowerment (EAGLE), AI3 Action Institute--Artificial Intelligence for American Indians (AI3 Action Institute), and the National Center for Native Training and Technical Assistance (NCNTTA).

In this document, the Federal Communications Commission (Commission or FCC) adopted a Second Report and Order that promotes the deployment of submarine cables while strengthening national security interests in connection with their modern capabilities. The Second Report and Order establishes a regulatory regime for licensing owners and/or operators of submarine line terminal equipment (SLTE), including granting a blanket license to certain current and future SLTE owners and/or operators that are

The Caribbean Fishery Management Council (CFMC) will hold the 190th public hybrid meeting to address the items contained in the tentative agenda included in the SUPPLEMENTARY INFORMATION.

By the authority vested in me as President by the Constitution and the laws of the United States of America, it is hereby ordered: Section 1. Policy. On March 27, 2025, I issued Executive Order 14253 (Restoring Truth and Sanity to American History). Pursuant to that order, the Domestic Policy Council recently completed a months-long review […] The post Restoring Trust in the Smithsonian Institution appeared first on The White House .

The post Investors Need Better Information About How Companies are Using AI appeared first on Partnership on AI .

GPT-5.6 improves AI efficiency across models, inference, and agentic workflows, helping deliver more useful intelligence per dollar.

A new field report shows how scientists use AI coding agents to modernize scientific computing, accelerating software development and discovery in genomics and beyond.

We’re announcing even more new capabilities in Managed Agents in Gemini API so developers can build reliable, production-ready agents.

For victims of targeted, sexualized attacks, searching for redress is not easy: they are blamed for posting pictures of them online, accused of overreacting by the police, and followed for life with the consequences of losing the control over their own body and image.

The post Steering AI’s Economic Impacts, Before They Arrive appeared first on Partnership on AI .

CT vision-language foundation models have demonstrated promising performance across downstream tasks, but are typically trained with whole-volume representations that dilute fine-grained anatomical signals. Fine-grained vision-language pre-training addresses this by aligning anatomy-level visual features with anatomy-specific text, but in doing so discards the global context that whole-volume models provide. Furthermore, existing fine-grained approaches train from scratch, making them computatio

There is growing interest in using large language models (LLMs) as low-cost proxies for resident attitudes in urban planning. Previous work shows that LLMs can predict average results of survey experiments, but less is known about whether they preserve the spatially anchored, identity-conditioned structure behind those averages, namely how support changes as a project approaches homes and how that response divides across tenure and partisan groups. We compared eight open-weight LLMs with 843 res

Scientific images are the core elements of presenting experimental conclusions, elaborating system architecture, and supporting comparative arguments in scientific papers. However, existing image quality assessment (IQA) methods are predominantly designed for natural photographs or AI-generated content, which cannot be directly applied to scientific papers. The few existing studies on scholarly charts remain confined to visual-surface comparisons, failing to verify caption alignment, citation re

As LLM agents increasingly depend on diverse external services such as search engines, databases, and connectors, agent harnesses face a fundamental tool-selection challenge: acquiring too few tools leaves the task under-informed, while too many adds cost, context load, and privacy exposure. Routers and retrievers can rank candidate tools by relevance, but a ranking alone does not determine how many are worth selecting. Existing approaches leave acquisition under heterogeneous costs unaddressed.

Fine-tuning is the dominant paradigm for specializing large language models (LLMs), yet it exposes a critical vulnerability: malicious data providers can embed harmful behaviors into downstream corpora, creating models that retain professional skills while violating human values on demand. Existing safety-realignment defenses often fail in practice due to three key limitations: they frequently cause catastrophic forgetting of specialized skills; their effectiveness collapses when the defender ca

Closed yes/no spatial benchmarks can reward a correct answer even when the image adds little support beyond no-image contexts. Under a fixed forced-choice interface, Visual Credit Audit (VCA) separates two estimands: whether the benchmark image gives the model's declared decision more support than text-only and blank controls, and whether the model responds to relation-specific visual evidence. The first audit is training- and label-free and does not require an answer flip. Applying labels yield