WhatsApp launches totally private'incognito' conversations with its AI chatbot WhatsApp has introduced private chats with its AI chatbot which not even the tech company will be able to read in a new incognito mode. It means neither the user nor the AI's responses will be monitored if the feature is activated, and past conversations will disappear from the chat for the user. Will Cathcart, the head of WhatsApp, said he felt people wanted to have private conversations with AI on sensitive subjects including health, relationships and finances and didn't want them to be accessible. But a cyber security expert has told the BBC this could lead to a lack of accountability for WhatsApp if things go wrong, as they would have no access to chat history. WhatsApp is owned by Meta, which also owns Instagram, Facebook and Messenger.

The Institute for Primary Facts has compiled more than 3.5 million pages of the Epstein files for public display at the newly opened Donald J. Trump and Jeffrey Epstein Memorial Reading Room. It's an early 2016 email thread between Jeffrey Epstein and a woman whose name is redacted by the Department of Justice . In the thread, Epstein asks the unidentified woman for a "naughty selfie" and later sends her a camera. In late February, he replies with a different ask: "Do you have any friends that might want to work for me?...I will give you money if you find someone willing to travel, 22-25, educated. The exchange carries extra resonance when you consider that Epstein is accused of sex trafficking minors, with the Department of Justice estimating that he had more than 1,200 potential victims.

HCCV datasets constructed through nonconsensual web scraping lack crucial metadata for comprehensive fairness and robustness evaluations. Current remedies are post hoc, lack persuasive justification for adoption, or fail to provide proper contextualization for appropriate application. Our research focuses on proactive, domain-specific recommendations, covering purpose, privacy and consent, and diversity, for curating HCCV evaluation datasets, addressing privacy and bias concerns. We adopt an ante hoc reflective perspective, drawing from current practices, guidelines, dataset withdrawals, and audits, to inform our considerations and recommendations.

Fans want to reclaim the music and myth of Michael Jackson in the new biopic while critics call for accountability. Still from, which opened April 24. Is truth determined by the size of the audience it reaches? If so, --a new film about the pop singer Michael Jackson that is on track to have the biggest-ever opening for a music biopic, with projected earnings of $70 million at the US box office, despite critics saying it sanitizes the reality of who Jackson actually was--intends to supplant the King of Pop as the apotheosis of artistic virtue. The film's release has sparked a familiar but newly intensified civil war online, between those eager to reclaim the music and myth of Jackson, and those who see any celebration of him as a failure of accountability.

Counterfactual explanations are emerging as an attractive option for providing recourse to individuals adversely impacted by algorithmic decisions. As they are deployed in critical applications (e.g. law enforcement, financial lending), it becomes important to ensure that we clearly understand the vulnerabilties of these methods and find ways to address them. However, there is little understanding of the vulnerabilities and shortcomings of counterfactual explanations. In this work, we introduce the first framework that describes the vulnerabilities of counterfactual explanations and shows how they can be manipulated. More specifically, we show counterfactual explanations may converge to drastically different counterfactuals under a small perturbation indicating they are not robust. Leveraging this insight, we introduce a novel objective to train seemingly fair models where counterfactual explanations find much lower cost recourse under a slight perturbation. We describe how these models can unfairly provide low-cost recourse for specific subgroups in the data while appearing fair to auditors. We perform experiments on loan and violent crime prediction data sets where certain subgroups achieve up to 20x lower cost recourse under the perturbation. These results raise concerns regarding the dependability of current counterfactual explanation techniques, which we hope will inspire investigations in robust counterfactual explanations.1

This happened as generative and agentic systems became essential in key sectors worldwide. This feature highlights the major AI ethics and policy developments of 2025, and concludes with a forward-looking perspective on the ethical and policy challenges likely to shape 2026.

In the past, MI datasets were frequently proprietary, confined to particular institutions, and stored in private repositories. In this particular setting, there is a pressing need for alternative models of data sharing, documentation, and governance. Within this context,theemergence ofCommunityContributed Platforms (CCPs) presented a potential for the public sharing of medical datasets.

However, finetuning often decreases accuracyand generalizability offoundation models[31]--a significant drawback as these models are commonly used for zero-shot tasks.