Critics though say the decision is linked to the changed political environment the CMA is now operating in. The government has instructed the country's regulators to suggest ways of stimulating economic growth. In January, the government removed the then chair of the CMA, Marcus Bokkerink, because it was unhappy with his response to that call. He was replaced on an interim basis by Doug Gurr, former boss of Amazon UK. "The CMA has sat on this decision for over a year, yet within just a few weeks of a former Amazon boss being installed as chair, it has decided everything was absolutely fine all along, nothing to see here," said Foxglove co-executive director Rosa Curling. "This is a bad sign that Big Tech has successfully convinced the prime minister to defang our competition regulator and let Big Tech gobble up the current generation of cutting-edge tech – just like they did the last one," she told the BBC.

Gladstone A.I. co-founders and CEOs Edouard Harris and Jeremie Harris explain the major role that A.I will play in national security and warfare on'The Will Cain Show.' Elon Musk met with members of the Senate DOGE caucus at the White House. MUSK'S MOVE BLOCKED: A California judge denied Elon Musk's move to halt OpenAI's efforts to convert it into a for-profit entity, saying in a ruling that the SpaceX and Tesla CEO hadn't met "the high burden required for a preliminary injunction." 'DOWNFALLS' OF AI: A federal judge has declined to impose sanctions on an attorney who submitted a brief that contained incorrect case citations and quotes generated by artificial intelligence. DEFEND YOUR DATA: Windows has always been a favorite target for hackers, but it seems they have now figured out how to actively target Macs as well. We've seen an alarming rise in malware affecting Mac computers, stealing personal data and cryptocurrency.

Elon Musk discovered a simple 30-second "fridge trick" that can reverse diabetes, but the discovery has spooked pharmaceutical companies so much they put a 78 million bounty on his head, forcing the Tesla CEO to flee the country. At least, that's what a collection of AI-generated Facebook ads claim. Facebook ads depicting deepfakes of Elon Musk and Fox News personalities claiming that the Tesla CEO has discovered the cure for diabetes have been circulating on the platform for weeks. The ads seem to be part of a wider scam that uses the deepfakes to sell unproven supplements. Engadget has identified scores of pages running versions of these ads since early February.

Hundreds of experts on international affairs believe World War III is inevitable and will likely start within the next 10 years. A new survey of 357 political strategists and foresight practitioners weighed in on the future of humanity, with four in 10 saying a major war involving powerhouses like the US, China or Russia will explode in 2035. By 2035, four in 10 global strategists (40.5%) predicted that a world war involving major nations like the United States, China, or Russia will break out. The majority of those who believe WWIII is coming said that it would likely involve nuclear weapons and battles in outer space. The most notable example pushing respondents to predict would likely be President Donald Trump establishing the US Space Force in 2019.

Democrat Vermont Sen. Bernie Sanders said the U.S. has become an "oligarchic society" while responding to President Donald Trump's address to a joint Congress Tuesday evening. "The Trump administration is not hiding it," Sanders said in a streamed response to Trump's address Tuesday. "The Trump administration is a government of the billionaire class by the billionaire class, and for the billionaire class. Notwithstanding some of their rhetoric, this is a government that could care less about ordinary Americans and the working families of our country. My friends, we are no longer moving toward oligarchy. We are living in an oligarchic society."

Taiwan is learning from companies in Ukraine that continue to operate during the country's fight against Russia, a senior Taiwan official said Wednesday, as the island speeds up contingency planning amid heightened Chinese threats. China claims democratically governed Taiwan as its own territory, despite the objection of the government in Taipei, and has ramped up its military pressure against the island in recent years, including holding several rounds of major war games. "We hope to learn from Ukraine's first-hand experiences -- how private companies helped build the resilience of its government and society during wartime," said a senior Taiwan security official, who requested anonymity due to the sensitivity of the matter.

Some companies are "stuck in neutral" in their approach to artificial intelligence, according to Microsoft's UK boss, who said a significant number of private and public sector organisations lack any formal AI strategy. A Microsoft survey of nearly 1,500 UK senior leaders across public and private sectors, as well as 1,440 employees, found that more than half of executives feel their organisation has no official AI plan. Roughly the same proportion report a growing gap in productivity – a measure of economic efficiency – between employees who use AI and those who do not. "Some organisations appear to be stuck in neutral, caught in the experimentation phase, rather than in the deployment [of AI]," said Darren Hardman, the tech company's UK chief executive. Microsoft, the biggest financial backer of the ChatGPT developer, OpenAI, has been pushing AI's deployment in the workplace through autonomous AI agents – tools that can carry out tasks without human intervention.

Efficient exploration is essential for intelligent systems interacting with their environment, but existing language models often fall short in scenarios that require strategic information gathering. In this paper, we present PAPRIKA, a fine-tuning approach that enables language models to develop general decision-making capabilities that are not confined to particular environments. By training on synthetic interaction data from different tasks that require diverse strategies, PAPRIKA teaches models to explore and adapt their behavior on a new task based on environment feedback in-context without more gradient updates. Experimental results show that models fine-tuned with PAPRIKA can effectively transfer their learned decision-making capabilities to entirely unseen tasks without additional training. Unlike traditional training, our approach's primary bottleneck lies in sampling useful interaction data instead of model updates. To improve sample efficiency, we propose a curriculum learning strategy that prioritizes sampling trajectories from tasks with high learning potential. These results suggest a promising path towards AI systems that can autonomously solve novel sequential decision-making problems that require interactions with the external world.

This paper describes the construction of a dataset and the evaluation of training methods to improve generative large language models' (LLMs) ability to answer queries on sensitive topics with a Neutral Point of View (NPOV), i.e., to provide significantly more informative, diverse and impartial answers. The dataset, the SHQ-NPOV dataset, comprises 300 high-quality, human-written quadruplets: a query on a sensitive topic, an answer, an NPOV rating, and a set of links to source texts elaborating the various points of view. The first key contribution of this paper is a new methodology to create such datasets through iterative rounds of human peer-critique and annotator training, which we release alongside the dataset. The second key contribution is the identification of a highly effective training regime for parameter-efficient reinforcement learning (PE-RL) to improve NPOV generation. We compare and extensively evaluate PE-RL and multiple baselines-including LoRA finetuning (a strong baseline), SFT and RLHF. PE-RL not only improves on overall NPOV quality compared to the strongest baseline ($97.06\%\rightarrow 99.08\%$), but also scores much higher on features linguists identify as key to separating good answers from the best answers ($60.25\%\rightarrow 85.21\%$ for presence of supportive details, $68.74\%\rightarrow 91.43\%$ for absence of oversimplification). A qualitative analysis corroborates this. Finally, our evaluation finds no statistical differences between results on topics that appear in the training dataset and those on separated evaluation topics, which provides strong evidence that our approach to training PE-RL exhibits very effective out of topic generalization.

Large Language Models (LLMs) are increasingly deployed across diverse contexts to support decision-making. While existing evaluations effectively probe latent model capabilities, they often overlook the impact of context framing on perceived rational decision-making. In this study, we introduce a novel evaluation framework that systematically varies evaluation instances across key features and procedurally generates vignettes to create highly varied scenarios. By analyzing decision-making patterns across different contexts with the same underlying game structure, we uncover significant contextual variability in LLM responses. Our findings demonstrate that this variability is largely predictable yet highly sensitive to framing effects. Our results underscore the need for dynamic, context-aware evaluation methodologies for real-world deployments.