The face-palm-worthy prompt injections against AI assistants continue. Today’s installment hits OpenAI’s Deep Research agent. Researchers recently devised an attack that plucked confidential information out of a user’s Gmail inbox and sent it to an attacker-controlled web server, with no interaction required on the part of the victim and no sign of exfiltration. Deep Research is a ChatGPT-integrated AI agent that OpenAI introduced earlier this year. As its name is…

Anthropic’s AI models could potentially help spies analyze classified documents, but the company draws the line at domestic surveillance. That restriction is reportedly making the Trump administration angry. On Tuesday, Semafor reported that Anthropic faces growing hostility from the Trump administration over the AI company’s restrictions on law enforcement uses of its Claude models. Two senior White House officials told the outlet that federal contractors working with agencies like the…

At multiple points over many years, Apple executives have taken great pains to point out that they think touchscreen Macs are a silly idea. But it remains one of those persistent Mac rumors that crops up over and over again every couple of years, from sources that are reliable enough that they shouldn’t be dismissed out of hand. Today’s contribution comes from supply chain analyst Ming Chi-Kuo, who usually has…

iOS 26 became publicly available this week, ushering in a new OS naming system and the software’s most overhauled look since 2013. It may take time to get used to the new “Liquid Glass” look, but it’s easier to appreciate the pared-down controls. Beyond a glassy, bubbly new design, the update’s flashiest new features also include new Apple Intelligence AI integration that varies in usefulness, from fluffy new Genmoji abilities…

Under new rules rolling out over the coming months, a small number of users will be required to leave some of their moderator posts so that they aren’t moderating more than five subreddits with 100,000 monthly visitors. Reddit said the change only affects “0.1 percent of our active mods” and will help enable “diverse perspectives and experiences.” But mods whom Ars Technica spoke with have different views. New limits on…

On Tuesday, OpenAI announced plans to develop an automated age-prediction system that will determine whether ChatGPT users are over or under 18, automatically directing younger users to a restricted version of the AI chatbot. The company also confirmed that parental controls will launch by the end of September. In a companion blog post, OpenAI CEO Sam Altman acknowledged the company is explicitly “prioritizing safety ahead of privacy and freedom for…

Google has so many products that it can be near-impossible to keep track. And yet, the company has rarely created desktop apps to go with those services. There are a handful, like Drive and Quick Share, but the company’s flagship product is only now coming to the desktop. The new Google app for Windows is available now, allowing you to search the web, Google Drive, and even your local files….

On Sunday, The New York Times reported that tens of millions of people are confessing secrets to AI chatbots trained on religious texts, with apps like Bible Chat reaching over 30 million downloads and Catholic app Hallow briefly topping Netflix, Instagram, and TikTok in Apple’s App Store. In China, people are using DeepSeek to try to decode their fortunes. In her report, Lauren Jackson examined “faith tech” apps that cost…

The last time Apple gave macOS a fresh design was in 2020’s macOS 11 Big Sur. That release was relatively light on new features and heavy on symbolism. Big Sur is also when Apple finally jettisoned the “10” in Mac OS X after two decades. More importantly, it was the first release installed on then-new Apple Silicon Macs, the culmination of a decade-plus of in-house chip design that began with…

When software engineer Joshua Fonseca recently connected the GameCube simulation classic Animal Crossing to a modern AI language model like the kind that powers ChatGPT, he decided to shake things up. By programming the AI to roleplay as villagers growing aware of their debt situation, and giving them a shared memory to track conversations, Fonseca orchestrated a scenario where the residents began to organize against their raccoon landlord. In Animal…