Bayesian experimental design (BED) is a principled framework for data-efficient design of sequential experiments. However, existing BED methods are unable to adapt to dynamic constraints inherent in real-world tasks due to budget limitations, varying costs, or physical constraints that restrict how designs evolve over time. In this paper, we introduce a novel approach to BED that enables constrained optimization of experimental designs by combining offline pre-training of an amortized policy and a posterior network with online multi-step lookahead planning using scenario trees. We empirically demonstrate that our method yields substantially more informative design sequences than existing methods across a range of constrained BED tasks, while incurring only a modest additional computational overhead.

Large language models (LLMs) are prone to hallucinations, i.e., statements unsupported by the input or training data, hindering reliable deployment. In parallel, numerous uncertainty estimation (UE) methods have been proposed to quantify model confidence and are often implicitly treated as proxies for model failure. However, the relationship between uncertainty and hallucinations remains insufficiently characterized. We present a systematic empirical study of the association between uncertainty estimators and hallucinations in LLMs. Rather than assuming this association, we evaluate directly when and to what extent it holds. We consider a diverse set of uncertainty estimators, including information-theoretic, sampling-based, and reflexive estimators, and examine their behavior across hallucination settings. Our experiments cover both intrinsic hallucinations (violations of input faithfulness) and extrinsic hallucinations (unsupported claims relative to training data), using four complementary benchmarks, including RAGTruth and HalluLens. We find that the association is highly variable and often weak, depending on the hallucination type and the LLM under evaluation. These results challenge the use of uncertainty as a direct signal of hallucination and clarify when it provides actionable information.

Privacy auditing aims to empirically assess privacy leakage in machine learning models using membership inference attacks (MIAs), and to derive lower bounds on differential privacy (DP) parameters. Recent one-run auditing methods address the high cost of standard approaches by relying on a single training run with multiple "canary" points whose inclusion or exclusion must be detected by the auditor. In this work, we study the problem of efficiently crafting canaries for one-run privacy auditing. Motivated by recent theoretical insights suggesting that interference between canaries contributes to weaker leakage estimates compared to multi-run methods, we propose to optimize canaries to be both highly detectable and minimally interfering. Our approach combines a greedy initialization based on influence functions with a bilevel optimization procedure that maximizes distinguishability while promoting diversity in embedding space, enabling the use of computationally efficient bilevel algorithms. Experiments show that our method achieves stronger privacy leakage estimates at a lower computational cost than existing canary crafting approaches.

Russia'relentlessly targeting' critical infrastructure and democracy, GCHQ says The UK is at a moment of consequence as Russia is relentlessly targeting critical infrastructure, the UK's largest spy agency will warn. GCHQ Director Anne Keast-Butler will set out threats facing the UK and the measures she believes need to be taken to confront them when she makes her inaugural public speech on Wednesday. Russia has been blamed for a string of espionage plots on British soil and, more recently, waging an undeclared'hybrid war' against the UK and other Nato countries. The Kremlin has denied the allegations. Keast-Butler says GCHQ is working tirelessly to fend off cyber attacks and counter what she calls reckless sabotage and assassination attempts.

When Pope Leo XIV presented his first encyclical on artificial intelligence at the Vatican on Monday, he invited Christopher Olah, cofounder of Anthropic, to speak. The move signaled an unprecedented alliance between the Catholic church and Silicon Valley. But to understand how this partnership came about, we need to go back to Anthropic's founding. Anthropic launched in 2021 after a group of OpenAI researchers, including Dario and Daniela Amodei, left to form a rival lab. They did so with a clear conviction: Artificial intelligence models were becoming too powerful to be developed exclusively according to the logic of competition and speed.

Summer has arrived, which means its vacation season--and there are plenty of travel tips to be found among the best movies on streaming this May. A bloody ballet battle royale in Budapest in Prime Video's a visit to the picturesque (and definitely not haunted) Dutch forests in Shudder's, or an action-packed trip to Japan courtesy of Netflix's, are just some of the locations sure to give you wanderlust this month. If you fancy something a bit more tropical, then look no further than on Hulu--although director Sam Raimi's twisty survival horror might have you thinking twice before turning on your out-of-office emails. And, if the rising temperatures are already too much, the Antarctic chill of John Carpenter's classic, and its 1950s inspiration,, are both landing on Criterion. Here are WIRED's picks of the best movies to watch right now.

What Pope Leo XIV's First Encyclical Says About the Power of AI In, the Pope decries the concentration of technological power in a few global players. Anthropic cofounder Chris Olah shakes hands with Pope Leo XIV ahead of the presentation of the first encyclical. An algorithm decides what we see, another filters what we read, and still others enter into the processes that govern work, information, and collective choices. But the text is not conceived as an exclusively technological reflection. Pope Leo XIV places the issue of AI within the tradition of the social doctrine of the Catholic Church and directly invokes--while updating it--the of Pope Leo XIII (published on May 15, 1891) in the year of its 135th anniversary.

PCWorld examines Pope Leo XIV's encyclical on AI, which emphasizes that artificial intelligence reflects creator biases and lacks genuine empathy or real-world experience. The Pope calls AI a "valuable tool that requires vigilance," advising users to adopt a more thoughtful, slower-paced approach when interacting with models like ChatGPT, Claude, and Gemini. This papal guidance encourages users to actively consider when, why, and what they ask AI systems, recognizing their limitations despite sophisticated responses. Delving into Pope Leo XIV's exhaustive treatise about humanity and AI, I was struck by a recurring theme: AI simulates fundamental human traits that it doesn't actually possess. For starters, AI lacks the grounding we humans get from our real-world experiences, Pope Leo noted in his first encyclical, which was released Monday by the Vatican. Yes, AI models like ChatGPT (or more specifically, GPT), Claude, and Gemini are trained on mountains of data that seemingly represent the entirety of human knowledge. But all that data is just that: data.

'Leave Kyiv': Why Russia's latest Ukraine threat is a major escalation Russia has urged foreigners to leave in Ukraine's capital Kyiv, and warned of more strikes on the city, suggesting a major escalation in its more-than-four-year-long war on Ukraine. In a statement issued on Monday, Russia's Ministry of Foreign Affairs said it planned to target "decision-making centres and command posts" and drone manufacturing facilities in the Ukrainian city in a series of strikes. The ministry's statement also urged Kyiv residents to avoid all military and administrative infrastructure facilities in the capital, which could be potential targets. A later statement said that Russian Foreign Minister Sergey Lavrov had advised US Secretary of State Marco Rubio of the plan and urged him to evacuate his embassy staff from Kyiv. Moscow said these planned strikes were in response to a drone strike on a student dormitory in Starobilsk, in the Russian-occupied Luhansk region of Ukraine, which killed at least 18 people. The threats come just days after Russian drone and rocket strikes on Kyiv on Saturday night killed at least four people and injured about 100 others.

AI-generated vloggers like Chloe VS History (left) and Nova VS History are, their creators say, 'taking an already-proven format and applying it to history' AI-generated vloggers like Chloe VS History (left) and Nova VS History are, their creators say, 'taking an already-proven format and applying it to history' The content creators behind channels like Chloe VS History are using AI tools to'bring history to life in a really visceral way' "I have just arrived in Tudor London, 1536," a young woman in a green puffer jacket tells the camera. "I'm going to check in at my room in the inn, get into the market. Then, later I am meeting the actual king - yep, Henry VIII - in person." On YouTube and other social platforms, users are flocking to watch AI-generated "history influencers", characters that vlog their travels to historical settings. One of the most popular channels is Chloe VS History, with more than 610,000 Instagram followers and 15m views on YouTube.