Mathematicians have never been so sought after by the world's richest people. At universities across the world, academics are seeing their colleagues mysteriously disappear and join private companies. Some of these companies are household names, like OpenAI and Google, but others are newly formed and just months old, hoping to capitalise on a moment in which mathematics is seen as the secret ingredient with which to improve artificial intelligence - which may in turn transform mathematics itself. "Last May, I was honestly kind of grieving for my scientific identity," says Ken Ono, who in 2025 went on leave from a professorship at the University of Virginia to join Axiom Math, a start-up aiming to build a maths-focused AI. Ono had been asked by a different company, called Epoch AI, to help craft a set of hard-to-solve maths problems that would test AI's problem-solving ability .

Does'federated unlearning' in AI improve data privacy, or create a new cybersecurity risk? As the capacity of artificial intelligence (AI) increases at an exponential rate, so do concerns about the privacy of user data . Increasingly, organizations around the world are adopting something called federated unlearning that enables AI training without centralizing sensitive data. This allows hospitals, banks and government agencies to collaborate while keeping data local -- an approach that's regarded as a major advance in privacy . Federated unlearning promises that user data can be removed from a trained AI system .

An Engineer's Post Protesting Laptop Surveillance Is Going Viral Inside Meta Meta employees in the US and UK are organizing against corporate software that tracks workers' keystrokes and mouse activity. Meta's decision to track employee keystrokes and mouse data is causing an uproar within the company. "Selfishly, I don't want my screen scraped because it feels like an invasion of my privacy," wrote an engineer in an internal post seen by nearly 20,000 coworkers this week. "But zooming out, I don't want to live in a world where humans--employees or otherwise--are exploited for their training data." The message aimed to rally support for a petition circulating inside the company since last Thursday that demands an end to what Meta calls the Model Capability Initiative.

In this piece, we reflect on AIES 2025, and outline the conversations and presentations from a discussion session on LLMs in the context of clinical usage and human rights. This is a crosspost from the latest issue of AI Matters, published by the ACM SIAGI. This year's conference on artificial intelligence, ethics and society (AIES) took place in the north of Madrid within the 180m-high tower block that forms the vertical campus of IE University. The event kicked off with a welcome from the chairs and organising committee members, with this opening session also featuring the conference best paper awards. Topics covered during the three-day event included mitigating bias, integrating AI into the workplace, evaluating LLMs in clinical settings, power dynamics in AI ecosystems, and dataset creation.

Why sovereignty over data and models is becoming a defining factor in enterprise AI success,as well as a prerequisite for forging safe agentic systems. When generative AI first moved from research labs into real-world business applications, enterprises made a tacit bargain: "Capability now, control later." Feed your proprietary data into third-party AI models, and you will get powerful results. But your data passes through systems you do not own, under governance you do not set. The protections you rely on are only as durable as the provider's next policy update. Now, with generative AI established in everyday business operations and sophisticated new agentic AI systems advancing every day, companies are reevaluating the terms of that deal.

The Download: deepfake porn's stolen bodies and AI sharing private numbers Plus: the US has approved Nvidia chip sales to 10 Chinese firms. When Jennifer got a research job in 2023, she ran her new professional headshot through a facial recognition program. She wanted to see whether it would pull up the porn videos she'd made more than a decade earlier. It did, but it also surfaced something she'd never seen before: one of her old videos, now featuring someone else's face on her body. Conversations about sexualized deepfakes usually focus on the people whose faces are inserted into explicit content without consent. But another group often gets ignored: the people whose bodies those faces are attached to.

The 10th International Workshop on Health Intelligence (W3PHIAI-26) celebrated a decade of bringing AI and health research together, building on a lineage that began with the AAAI-W3PHI workshops focused on population health (2014-2016), the AAAI-HIAI workshops focused on personalized health (2013-2016), and the subsequent joint W3PHIAI workshops held annually from 2017 through 2025. Over this decade, the series has produced hundreds of talks and high-impact publications that have collectively received thousands of citations, shaping the research agenda in both population health intelligence and personalized healthcare AI. This year's special theme, "Foundation Models and AI Agents," reflected the field's rapidly evolving frontier: the emergence of autonomous and semi-autonomous AI systems reshaping clinical workflows, patient management, health system operations, and public health surveillance. Day 1 of the workshop focused on medical imaging and the translation of AI for clinical ...

A major bottleneck in characterizing the failure modes of generative AI systems is the cost and time of annotation and evaluation. Consequently, adaptive testing paradigms have gained popularity, where one opportunistically decides which cases and how many to annotate based on past results. While this framework is highly practical, its extreme flexibility makes it difficult to draw statistically rigorous conclusions, as it violates classical assumptions: the number of observations is typically limited (often 10 to 50 cases) and decisions regarding sampling and stopping are made in the midst of data collection rather than based a pre-specified rule. To characterize what statistical inferences can be drawn from highly adaptive audits, we introduce a hypothesis testing framework from two 'dueling' perspectives: (i) the model's null that asserts there is no failure mode with performance below a target threshold versus (ii) the auditor's null that asserts they have a sampling strategy that will uncover a failure mode. Leveraging Safe Anytime-Valid Inference (SAVI), we formalize the auditor as conducting 'testing by betting', which translates into simultaneous e-processes for testing the dueling null hypotheses. Furthermore, if the auditor is sufficiently powerful, we prove that these two hypotheses are asymptotically inverses of each other, in that passage of a stringent audit does in fact certify the AI system as being globally robust. Empirically, we demonstrate that our proposed testing procedures maintain anytime-valid type-I error control, outperform pre-specified testing methods, and can reach statistically rigorous conclusions sometimes with as few as 20 observations.

To stop children from bypassing its age checks, Meta is revamping its age-verification tools with an AI system that analyzes images and videos for "visual cues," such as height and bone structure. Meta is beefing up its age-verification mechanisms with an AI system that analyzes images and videos on Instagram and Facebook for "visual cues," such as height and bone structure, to identify and delete accounts of users under the age of 13. The company announced the move amid a wave of cases in which hundreds of children have managed to evade social network access restrictions, even through simple tricks such as drawing on a mustache. The new approach is part of a series of measures Meta adopted as part of an AI-based security strategy designed to correct the limitations of traditional methods, which rely heavily on self-reported age. With this change, the company seeks to reduce the ease with which minors access platforms that, in theory, are restricted to them.

New research suggests that reliance on AI assistants can have a negative impact on people's ability to think and problem solve. Using AI chatbots for even just for 10 minutes may have a shockingly negative impact on people's ability to think and problem-solve, according to a new study from researchers at Carnegie Mellon, MIT, Oxford, and UCLA. Researchers tasked people with solving various problems, including simple fractions and reading comprehension, through an online platform that paid them for their work. They conducted three experiments, each involving several hundred people. Some participants were given access to an AI assistant capable of solving the problem autonomously.