AI models such as ChatGPT routinely misrepresent news events, providing faulty responses to questions almost half the time, a study has found. The study published on Wednesday by the European Broadcasting Union (EBU) and the BBC assessed the accuracy of more than 2,700 responses given by OpenAI's ChatGPT, Google's Gemini, Microsoft's Copilot, and Perplexity. Overall, 45 percent of responses had at least one "significant" issue, according to the research. Sourcing was the most common problem, with 31 percent of responses including information not supported by the cited source, or incorrect or unverifiable attribution, among other issues. A lack of accuracy was the next biggest contributor to faulty answers, affecting 20 percent of responses, followed by the absence of appropriate context, with 14 percent.

Author Philip Pullman calls on government to act over'wicked' AI scraping Writers whose work has been scraped don't get compensation or recognition, something authors including Kate Mosse and Richard Osman have criticised, saying it could destroy growth in creative fields and amount to theft. Sir Philip, author of the hugely popular novels about Lyra Silvertongue, the heroine of His Dark Materials and The Book of Dust trilogies, thinks writers should be compensated. They can do what they like with my work if they pay me for it, he told the BBC's culture editor Katie Razzall. The Department for Culture, Media and Sport has been contacted for a response to Sir Philip's comments. Sir Philip said: As far as I know everybody's work has been stolen, scraped like a trawler... at the bottom of the sea. You name it, it's all killed.

How successful is Ukraine's'gas war' against Russia? How will Putin travel to Hungary with an ICC arrest warrant? How much of Europe's oil still comes from Russia? A "massive" Russian attack killed four people and injured seven in the town of Novhorod-Siverskyi, in Ukraine's Chernihiv region, Governor Viacheslav Chaus wrote in a post on Telegram. Chaus said that Russian forces launched about 20 Shahed drones in the attack and that there was "a lot of destruction in the city".

ChatGPT-maker OpenAI has unveiled an artificial intelligence-powered web browser to challenge competitors like Google, which operates Chrome, the most popular browser in the world. ChatGPT Atlas does away with the address bar that is a key feature in search, with boss Sam Altman saying it was built around ChatGPT as the company made the new browser available on Tuesday on Apple's MacOS operating system. The arrival of Atlas comes as OpenAI seeks new ways to monetise its massive bet on artificial intelligence (AI) and capitalise on its growing user base. OpenAI said Atlas would also offer a paid agent mode that conducts searches on its own for users of its popular chatbot. The agent mode feature will be available only to paying ChatGPT subscribers.

High-dimensional non-convex loss landscapes play a central role in the theory of Machine Learning. Gaining insight into how these landscapes interact with gradient-based optimization methods, even in relatively simple models, can shed light on this enigmatic feature of neural networks. In this work, we will focus on a prototypical simple learning problem, which generalizes the Phase Retrieval inference problem by allowing the exploration of overparametrized settings. Using techniques from field theory, we analyze the spectrum of the Hessian at initialization and identify a Baik-Ben Arous-Péché (BBP) transition in the amount of data that separates regimes where the initialization is informative or uninformative about a planted signal of a teacher-student setup. Crucially, we demonstrate how overparameterization can bend the loss landscape, shifting the transition point, even reaching the information-theoretic weak-recovery threshold in the large overparameterization limit, while also altering its qualitative nature. We distinguish between continuous and discontinuous BBP transitions and support our analytical predictions with simulations, examining how they compare to the finite-N behavior. In the case of discontinuous BBP transitions strong finite-N corrections allow the retrieval of information at a signal-to-noise ratio (SNR) smaller than the predicted BBP transition. In these cases we provide estimates for a new lower SNR threshold that marks the point at which initialization becomes entirely uninformative.

Large language models have become the latest trend in natural language processing, heavily featuring in the digital tools we use every day. However, their replies often reflect a narrow cultural viewpoint that overlooks the diversity of global users. This missing capability could be referred to as cultural reasoning, which we define here as the capacity of a model to recognise culture-specific knowledge values and social norms, and to adjust its output so that it aligns with the expectations of individual users. Because culture shapes interpretation, emotional resonance, and acceptable behaviour, cultural reasoning is essential for identity-aware AI. When this capacity is limited or absent, models can sustain stereotypes, ignore minority perspectives, erode trust, and perpetuate hate. Recent empirical studies strongly suggest that current models default to Western norms when judging moral dilemmas, interpreting idioms, or offering advice, and that fine-tuning on survey data only partly reduces this tendency. The present evaluation methods mainly report static accuracy scores and thus fail to capture adaptive reasoning in context. Although broader datasets can help, they cannot alone ensure genuine cultural competence. Therefore, we argue that cultural reasoning must be treated as a foundational capability alongside factual accuracy and linguistic coherence. By clarifying the concept and outlining initial directions for its assessment, a foundation is laid for future systems to be able to respond with greater sensitivity to the complex fabric of human culture.

We identify semantically coherent, context-consistent network components in large language models (LLMs) using coactivation of sparse autoencoder (SAE) features collected from just a handful of prompts. Focusing on concept-relation prediction tasks, we show that ablating these components for concepts (e.g., countries and words) and relations (e.g., capital city and translation language) changes model outputs in predictable ways, while amplifying these components induces counterfactual responses. Notably, composing relation and concept components yields compound counterfactual outputs. Further analysis reveals that while most concept components emerge from the very first layer, more abstract relation components are concentrated in later layers. Lastly, we show that extracted components more comprehensively capture concepts and relations than individual features while maintaining specificity. Overall, our findings suggest a modular organization of knowledge accessed through compositional operations, and advance methods for efficient, targeted LLM manipulation.

The £5.30 orange juice that tells the story of why supermarket prices are sky high There has been more than a bitter twang in the glasses at British breakfast tables. Only five years ago, a typical supermarket own-label carton of orange juice could be bought for 76p for 1 litre. One colleague was outraged to be sent a bill for £9 for a glass of hangover-busting orange juice and lemonade at an unassuming little restaurant in Kent. Asked why so much, she was told that the orange juice - albeit freshly squeezed - accounted for £5.30 of the price. Yet as costs have surged, the taste is changing too, with certain manufacturers substituting oranges for mandarins to cut costs.

The head of global theme park giant Merlin Entertainments says its biggest competition is people choosing to stay at home on their phones and other devices. Fiona Eastwood says a day out at one of its UK attractions - which include Legoland, Thorpe Park and Alton Towers - was the perfect antidote for spiralling screen time. In a wide-ranging BBC Big Boss Interview, the chief executive reflected on challenges in the forthcoming Budget, big brand partnerships, and how its customers were responding to cost-of-living pressures. Eastwood also highlighted the importance of seasonal attractions to its customers - with Halloween now rivalling its summer season in driving profits at some attractions. Having been in the job since February, Eastwood has taken over at a time when her industry is facing challenges from a dip in consumer confidence.

The Ukrainian city of Chernihiv is in total blackout following what the authorities describe as a massive assault by Russian missiles and drones, with hundreds of thousands of people affected. Across the wider Chernihiv region, four people are reported to have been killed as residential neighbourhoods were struck in the town of Novhorod-Siverskyi. Ten others were injured, including a 10-year-old girl. The country's most northerly region is the latest to be hit in an intensifying series of attacks on civilian infrastructure as Russia targets energy supplies, the rail network, homes and businesses in its full-scale invasion of Ukraine. I personally heard the drones flying overhead, 55-year-old Oleksandr Babich said.