Skip to content
Tech News
clear
Topics: Today This Week This Month This Year

ICE Stopped Tracking Miscarriages in Detention After Recording 18 Cases

A Guardian investigation found ICE documented 18 miscarriages among detained women in the first nine months of Trump's second term, but stopped keeping records after October 2025, leaving an unknown number of subsequent losses uncounted. One detainee, an asylum seeker named Anabell, described being denied prenatal care and a doctor's visit for weeks despite bleeding, before finally being taken to an emergency room shackled to a bed, where she learned she had lost her pregnancy.

Google Messages extends chat themes to standard SMS conversations

Google is quietly expanding its chat themes feature in Google Messages beyond RCS conversations to include standard SMS texts, according to a Reddit user's discovery confirmed on a Pixel 6a. Users can access the feature through the three-dot menu or contact info page, though Google hasn't published official documentation about the rollout.

BeanShell 3.0 Development Resumes After Long Hiatus

Work has restarted on BeanShell, the lightweight scripting engine for Java, aiming for a full 3.0 release following an extended period of inactivity. The updated version is planned to modernize the codebase and add long-requested features, including support for lambda expressions.

Second Circuit rules border agents can search phones without a warrant

The Second Circuit Court of Appeals ruled that U.S. border agents may search travelers' electronic devices without any individualized suspicion, rejecting arguments for a warrant requirement. The decision came in United States v. Alisigwe, a case in which prosecutors used evidence from two warrantless cellphone searches conducted at the border. The Knight First Amendment Institute and the Reporters Committee for Freedom of the Press had filed a brief urging the court to require warrants, citing First and Fourth Amendment concerns.

Security firm Hacktron used Anthropic's Claude to breach OpenAI employee accounts

A three-person team at startup Hacktron AI used Anthropic's Claude AI model to chain two vulnerabilities together, gaining access to several OpenAI employees' ChatGPT accounts and internal systems through OpenAI's bug-bounty program. The entry point traced back to a flaw in Discourse, the third-party forum software OpenAI uses, triggered through a routine image upload. OpenAI paid Hacktron $6,500 and says it has since patched the vulnerabilities.

Tsinghua engineer Ning Xu wins Nikon Small World in Motion for cilia video

Ning Xu, an optical engineer at Tsinghua University in Beijing, won Nikon's 2023 Small World in Motion competition with a super-resolution video capturing the abnormal beating of airway cilia in a child with primary ciliary dyskinesia (PCD). His entry topped 346 submissions from 40 countries in the contest, which is a video-focused offshoot of Nikon's long-running Small World Photomicrography Competition.

Truck driver's homemade NYC model, viral on TikTok, now exhibited at city museum

Joe Macken, a truck driver from New York, spent over 20 years building an intricate scale model of the city from balsa wood and foam board in his basement, eventually creating more than 800,000 structures. After his daughter posted videos of the project on TikTok and they drew over 10 million views in a week, the Museum of the City of New York invited him to display it. The exhibit, titled 'He Built This City,' opened in February and runs through October 12.

White-hat hackers breach OpenAI's internal codebase via Discourse forum flaw

Cybersecurity firm Hacktron AI disclosed that its researchers exploited a chained vulnerability—an outdated image-processing library on OpenAI's Discourse forum combined with an SSO misconfiguration—to hijack employee ChatGPT and Codex accounts and reach OpenAI's private code repository. To prove the breach, they submitted a harmless pull request to OpenAI's internal monorepo before reporting the flaws through OpenAI's bug bounty program. OpenAI patched the issue within 14 hours and paid the team a $6,500 bounty.

Researchers bake replica scroll to test lead-based ink detection for Herculaneum texts

Scientists studying the carbonized Herculaneum scrolls have created an artificial scroll and charred it to mimic the effects of the Vesuvius eruption, aiming to test whether lead traces in ancient ink can help reveal hidden text. The experiment is meant to validate imaging techniques before they are applied to the fragile, irreplaceable originals.

Chipotle, Taco Bell, IHOP Turn Customer TikTok Hacks Into Menu Items

Major fast-food chains are officially adopting menu items first popularized by customers on social media. IHOP revived its Stuffed French Toast after fan requests and rolled out Dubai Chocolate Pancakes nationwide following a viral test run, while Chipotle launched Chili Lime Chips inspired by customers crushing chips into their bowls. Taco Bell's new 'Fan Style' feature lets loyalty members build and share custom orders, following similar moves by McDonald's and Burger King.

Today's top topics: openai anthropic apple ai safety iphone 18 pro dario amodei ios 27 artificial intelligence google nvidia
View all today's topics →