An Oxford-led study put millions of questions about towns, cities and countries to ChatGPT. Wealthier regions repeatedly came out ahead, raising questions about how AI can influence our picture of
Read articleResearchers found that an AI-generated image may be impossible to trace convincingly to one artist simply by testing how much the result changes without their work. That complicates easy claims about creative credit.
Read article
An AI system defeated elite human Stratego players despite never seeing the identities of many pieces on the
Europe has started applying transparency rules to AI-generated content. The obligations differ for chatbots, synthetic images, realistic deepfakes
AI can draw a spectacular dragon-shaped mug and still forget that a cup needs to hold liquid. MIT
Researchers want to bring scattered bus and train information together so passengers get clearer updates when journeys go
An Oxford-led study put millions of questions about towns, cities and countries to ChatGPT. Wealthier regions repeatedly came out ahead, raising questions about how AI can influence our picture of places we have never visited.
Read article
MIT researchers have shown a drone navigating around moving obstacles in surroundings it did not know in advance.
Anthropic has launched Claude Haiku 5.5 for quick, repetitive AI tasks at sharply lower prices than its predecessor.
ChatGPT can now build a working calculator, chart or other interactive element in the middle of a reply.
Google has opened an AI-fake checking tool to everyone, allowing people to test images, videos and audio for
Atlassian and OpenAI are expanding their partnership so GPT-6 models can work with Rovo and the Teamwork Graph, bringing company context into planning, coding and agent workflows.
Read article
Pinterest Beauty Guides analyse hair and nail Pins, then generate salon language, step-by-step guidance and product suggestions. The
Anthropic has expanded its Cyber Verification Program so approved security teams can use more capable Claude cyber functions
OpenAI has added native audio uploads to ChatGPT for paid users. Meetings, interviews and lectures can be transcribed,
Researchers built a soft robot that stores energy from daily heating and cooling as pneumatic pressure. One outdoor
A flexible tensegrity robot can tumble over rough terrain, descend a two-metre cliff and survive a 5.7-metre drop onto asphalt. Its body bends, senses and recovers instead of treating every fall as failure.
Read articleA meta-analysis found no single productivity effect from robot presence. Interactive robots were linked with faster task performance,
A Stanford study estimates that the mean amount Americans would need to give up generative AI for a
In a randomised experiment, generative AI sharply narrowed the performance gap between adults with different education levels. The
A 41-country Stanford study finds AI-adopting firms are shifting their workforce mix towards senior staff. The risk may
AI safety careers extend beyond research scientists. Engineering, evaluations, cybersecurity, policy, governance and operations all offer routes into work focused on understanding and reducing AI risk.
Read articleThere is no single route into AI. A useful career plan starts by choosing whether you want to
AI law careers now span legal practice, privacy, governance, product work and lawtech. The strongest route combines real
OpenAI says a coordinated model-extraction campaign generated 16,000 suspicious requests across two July days. The company says attackers
Google is giving Gemini 4 Argon to a limited group of cyber defenders before a wider release. The
Barclays says Claude now helps more than 16,000 employees find internal answers and processes about 120,000 market emails a day. The bank also wants half its developers using Claude Code by year end.
Read articleA Nature study turns scientific papers into AI agents that can run methods, answer questions and collaborate. In
MIT researchers built a robotic optics lab that assembled a working laser from loose components in 30 minutes.
A Cornell-led project is developing orchard robots that could pollinate flowers, thin fruit, weed and harvest apples. The
Albertsons is expanding ChatGPT from a Safeway shopping experience into selected internal teams and retail workflows. The October
Stanford researchers have built oversight frameworks that teach AI agents when to act alone and when to ask a human. The goal is useful autonomy without forcing people to approve every routine step.
Read articleFood companies can see how pooled data could help AI spot rare safety risks earlier. Cornell researchers found
MIT researchers argue that one shared hiring algorithm does not automatically harm job seekers. Their models suggest the
California’s attorney general has subpoenaed OpenAI as part of an investigation into cybersecurity incidents and model risks. The
A Stanford-led audit found AI models scored differently through developer APIs and consumer chat interfaces, suggesting benchmark results
MIT researchers used a two-stage AI controller to make an insect-scale flying robot 447% faster, with 255% greater acceleration and 10 consecutive flips in 11 seconds.
Read articleA study of 609 young adults found AI transparency changed the link between rational reactions to in-feed adverts
Six experiments found people were more willing to behave dishonestly towards AI agents, partly because they expected less
A study of 847 female online shoppers linked AI nudges and social cues with addictive shopping behaviour, while
A peer-reviewed comparison found AI financial advice showed less projection bias than human advisers but tended to recommend
University of Maryland researchers taught a robot system to learn how objects should move from ordinary videos, then translate those motion patterns into commands for a specific robot.
Read article
Nvidia’s new 64GB DGX Spark starts at $4,999 and can link two boxes into a 128GB local AI
Anthropic is putting $100 million behind a programme to train 10,000 engineers who can turn Claude projects into
HardFlow steers flow-matching generative models so their final outputs satisfy formal hard constraints. The method was tested on
Carnegie Mellon researchers developed LAMP, a planning framework that lets teams of robots reason about the object's route
NielsenIQ says 51% of US consumers in its recurring tracker used at least one AI-powered tool to support shopping in the previous month. The figure covers recommendations and assistants, not simply autonomous purchases.
Read article
Anthropic is interviewing Claude users about what they want from AI and letting participants publish the complete transcript.
Microscopic robots equipped with temperature sensors and artificial cilia can coordinate their movements to pump fluid and reshape
Researchers analysed real ten-minute speed dates using facial, audio and language signals. Their models found suggestive patterns, but
A survey of 360 university students in China found that those who saw generative AI as more empowering