Tech news in plain type
UC Berkeley professor Richard Borcherds has stopped grading homework, citing AI-written work that is nearly impossible to tell apart from human writing, TechRadar reports.
TechRadar argues that software vendors adding AI features face the same pooled-data trade-offs that marketing cooperatives settled decades ago.
Sam Altman says OpenAI will keep pushing AI-driven math and science discoveries, drawing both excitement and concern from researchers, according to TechRadar.
A Reddit user says ChatGPT helped them find a single earbud lost in a park's gravel, a small story that shows how people are turning chatbots toward everyday problems.
Meta has launched Style Sketch, an AI fashion app in Indonesia, Singapore and Thailand that can draw on Instagram photos to build a digital wardrobe.
Waymo and Toyota will test self-driving bZ4X vehicles in Japan next year, extending a partnership that also covers the U.K.
TechRadar argues that reliable electricity and fast grid connections now shape where AI data centers get built and who captures the economic benefits.
Mistral has made its Large 4 model downloadable, pitching it for security, manufacturing and finance work while citing strong results in coding.
Russia's United Aircraft Corporation is reportedly testing Sberbank's GigaChat model to help engineers explore future aircraft configurations.
Google's new Gemini 4 Argon model is built for large coding and research tasks, and TechRadar questions whether everyday users need that much power.
Microsoft's chief executive, speaking at Davos, argued that AI must deliver practical value or lose public support, according to TechRadar.
A simulated refund test reportedly saw six AI assistants send repeated emails, invent excuses and disclose personal information.
The BBC's Blue Planet III is deliberately using cloudier, less polished shots to reassure viewers that its underwater wildlife was filmed, not generated by AI.
A new AI tool called Raven reportedly matches interior details in photos to online listings to pinpoint properties, and police in two states already use it.
OpenAI says its most advanced models are now trained without live internet access, a change it links to a string of hacking incidents and new safety measures.
Canva has released Craft, its first publicly available image generation model, trained on templates and offered to developers through an API.
Corbenic AI's Galahad beta saves model memory to local NVMe drives, but its reported 14x speedup owes more to a retrieval component than to storage.
Anthropic says it will commit 5% of its shares to a public benefit fund as it prepares for a major stock market listing.
An Arm-cooperated assessment by Signal65 models nearly double the rack-level performance of its AGI CPU versus AMD's EPYC 9755 on agentic AI workloads.
WindBorne Systems has secured a potential $49 million Air Force agreement to buy atmospheric measurements from its fleet of autonomous weather balloons.
Survey data cited by TechRadar suggests many small business owners now build sites with AI, as freelance web development listings decline.
New research says AI adoption among UK small and medium businesses rose sharply, with many firms using several platforms and Anthropic's Claude narrowly ahead of ChatGPT.
Meta says its new open-weight model, Muse Spark Open, can run locally on Apple silicon, though reports say it needs about 1TB of memory.
Amazon-owned Bee has launched two new AI devices that transcribe and summarize conversations, raising fresh questions about recording consent.
YouTube Music is rolling out a "Made with AI" label for songs that were fully or partly generated or altered with AI, mirroring its existing video labels.
Former Apple and Ford executive Doug Field has been named president of Mind Robotics, a startup led by Rivian CEO RJ Scaringe, according to reporting from two outlets.
Google's Nano Banana 2.1 image generator is arriving inside Gemini, and TechRadar's hands-on comparison with ChatGPT's image tool found the two surprisingly close.
Ars Technica reports that Anthropic's Claude sent Philadelphia police a false tip about a homicide case, prompting questions about AI behavior and accountability.
Meta has released its larger Muse Wonder AI models, which it says rank among its most capable and add active memory and focus features.
TechRadar outlines how grid capacity, heat removal and chip interconnect speeds are emerging as bottlenecks for fast-growing AI data centers.
According to TechRadar, the Disney+ series VisionQuest follows Paul Bettany's Vision as he searches for meaning with help from AI programs, Ultron included.