Nine people have been killed and dozens more wounded in a Russian attack on the western city of Ternopil, Ukraine's president Volodymyr Zelensky has said. Nine-storey blocks of flats were hit in the strikes, as Russia fired more than 470 drones and 47 missiles at Ukraine overnight in a brazen attack, Zelensky said. Three districts of Ukraine's second city, Kharkiv, were also hit by a massive drone attack which injured more than 30 people, including children. Photos posted online showed buildings and cars ablaze. Power cuts are affecting a number of regions across the country, Ukraine's energy ministry said.

Prime Minister Donald Tusk calls the Warsaw-Lublin railway explosion an "unprecedented act of sabotage" as Poland accuses Russian intelligence of the attack.

Is the fall of Pokrovsk inevitable? Is Trump losing patience with Putin? A Russian missile strike on the eastern Ukrainian city of Balakliia killed three people and wounded 10, including three children, a regional military official in the Kharkiv region said on Telegram on Monday. At least two people were killed and three were injured in Russian shelling of the Nikopol district in Ukraine's Dnipropetrovsk region, Vladyslav Haivanenko, the acting head of the Dnipropetrovsk Regional Military Administration, wrote on Facebook. Russian troops captured three villages across three Ukrainian regions, the RIA news agency cited the Russian Ministry of Defence as saying on Monday.

Adversarial optimization algorithms that explicitly search for flaws in agents' policies have been successfully applied to finding robust and diverse policies in multi-agent settings. However, the success of adversarial optimization has been largely limited to zero-sum settings because its naive application in cooperative settings leads to a critical failure mode: agents are irrationally incentivized to self-sabotage, blocking the completion of tasks and halting further learning. To address this, we introduce Rationality-preserving Policy Optimization (RPO), a formalism for adversarial optimization that avoids self-sabotage by ensuring agents remain rational--that is, their policies are optimal with respect to some possible partner policy. To solve RPO, we develop Rational Policy Gradient (RPG), which trains agents to maximize their own reward in a modified version of the original game in which we use opponent shaping techniques to optimize the adversarial objective. RPG enables us to extend a variety of existing adversarial optimization algorithms that, no longer subject to the limitations of self-sabotage, can find adversarial examples, improve robustness and adaptability, and learn diverse policies. We empirically validate that our approach achieves strong performance in several popular cooperative and general-sum environments. Our project page can be found at https://rational-policy-gradient.github.io.

Despite rapid growth in multimodal large language models (MLLMs), their reasoning traces remain opaque: it is often unclear which modality drives a prediction, how conflicts are resolved, or when one stream dominates. In this paper, we introduce modality sabotage, a diagnostic failure mode in which a high-confidence unimodal error overrides other evidence and misleads the fused result. To analyze such dynamics, we propose a lightweight, model-agnostic evaluation layer that treats each modality as an agent, producing candidate labels and a brief self-assessment used for auditing. A simple fusion mechanism aggregates these outputs, exposing contributors (modalities supporting correct outcomes) and saboteurs (modalities that mislead). Applying our diagnostic layer in a case study on multimodal emotion recognition benchmarks with foundation models revealed systematic reliability profiles, providing insight into whether failures may arise from dataset artifacts or model limitations. More broadly, our framework offers a diagnostic scaffold for multimodal reasoning, supporting principled auditing of fusion dynamics and informing possible interventions.

Fox News contributors Katie Pavlich and Miranda Devine discuss how President Donald Trump could be the one to bring an end to the Russia-Ukraine war on'Hannity.' Rep. Marjorie Taylor Greene, R-Ga., late Thursday took shots at Ukrainian President Volodymyr Zelenskyy, accusing him of trying to sabotage Friday's highly anticipated peace talks between President Donald Trump and Russian President Vladimir Putin by launching drone strikes on Russia. Greene responded to a post on X from the account, "Open Source Intel," which reported that Ukraine had in recent hours launched "one of the largest" drone attacks on Russia. "On the eve of the historic peace talks between President Trump and President Putin, Zelensky does this," the Republican lawmaker wrote. "Zelensky doesn't want peace and obviously is trying to sabotage President Trump's heroic efforts to end the war in Ukraine. Fox News Digital has reached out to the Ukrainian embassy, seeking a response to Greene's post. Rep. Marjorie Taylor Greene, R-Ga., accused Ukrainian President Zelenskyy of trying to sabotage peace talks between President Trump and Russian President Putin by launching drone strikes on Russia. Ukraine launched multiple drone strikes into Russia overnight Thursday, damaging several apartment buildings in the southern city of Rostov-on-Don and injuring more than a dozen civilians, according to acting governor of the region, Yuri Slyusar. Two of those wounded were hospitalized in serious condition, he said. The Ukrainian strikes came after Russian strikes in Ukraine's Sumy region overnight Wednesday, resulting in multiple injuries, including a 7-year-old girl, per officials. Local officials also accused Ukraine of launching a drone strike in Belgorod that injured three people, and another that struck a car in the village of Pristen that killed at least one individual. Ukrainian President Volodymyr Zelenskyy will not attend the summit in Alaska on Friday between President Donald Trump and Russian President Vladimir Putin. Despite the violence, Trump and Putin are scheduled to meet in Anchorage, Alaska, on Friday for a high-stakes summit on the future of the Ukraine war. The meeting will mark Putin's first visit to the U.S. since 2015 and the first U.S.-Russia summit since June 2021. President Donald Trump will meet with Russian President Vladimir Putin in Alaska on Aug. 15, 2025. Putin praised the U.S. on Thursday for making "sincere efforts" to end the war between Russia and Ukraine, which has been raging since early 2022. Appearing on television, the Russian president said the U.S. was "making, in my opinion, quite energetic and sincere efforts to stop hostilities, stop the crisis and reach agreements that are of interest to all parties involved in this conflict." Zelenskyy accused Russia of not being sincere in its intention to wind down the war. "This war must be ended.

Philosophy professor Dr. Susan Schneider joins'Fox & Friends First' to discuss the impact of artificial intelligence on students' performance in the classroom. Staying awake to watch "just one more episode" is a classic excuse for delaying bedtime. And with popular shows like Peacock's "Love Island" airing almost every night as the drama unfolds live, there's more pressure to finish the latest episode and to engage in conversation with others the next day. In addition to making us sleepier in the morning, staying awake to watch TV is not good for the brain, according to Daniel Amen, a psychiatrist, brain imaging doctor and founder of Amen Clinics in California. "'I just have to watch the last episode' of whatever show you're watching, and you end up cutting out half an hour or an hour of sleep," he said in an interview with Fox News Digital.

Recent work has demonstrated the plausibility of frontier AI models scheming -- knowingly and covertly pursuing an objective misaligned with its developer's intentions. Such behavior could be very hard to detect, and if present in future advanced systems, could pose severe loss of control risk. It is therefore important for AI developers to rule out harm from scheming prior to model deployment. In this paper, we present a suite of scheming reasoning evaluations measuring two types of reasoning capabilities that we believe are prerequisites for successful scheming: First, we propose five evaluations of ability to reason about and circumvent oversight (stealth). Second, we present eleven evaluations for measuring a model's ability to instrumentally reason about itself, its environment and its deployment (situational awareness). We demonstrate how these evaluations can be used as part of a scheming inability safety case: a model that does not succeed on these evaluations is almost certainly incapable of causing severe harm via scheming in real deployment. We run our evaluations on current frontier models and find that none of them show concerning levels of either situational awareness or stealth.

This paper presents an adversary model and a simulation framework specifically tailored for analyzing attacks on distributed systems composed of multiple distributed protocols, with a focus on assessing the security of blockchain networks. Our model classifies and constrains adversarial actions based on the assumptions of the target protocols, defined by failure models, communication models, and the fault tolerance thresholds of Byzantine Fault Tolerant (BFT) protocols. The goal is to study not only the intended effects of adversarial strategies but also their unintended side effects on critical system properties. We apply this framework to analyze fairness properties in a Hyperledger Fabric (HF) blockchain network. Our focus is on novel fairness attacks that involve coordinated adversarial actions across various HF services. Simulations show that even a constrained adversary can violate fairness with respect to specific clients (client fairness) and impact related guarantees (order fairness), which relate the reception order of transactions to their final order in the blockchain. This paper significantly extends our previous work by introducing and evaluating a mitigation mechanism specifically designed to counter transaction reordering attacks. We implement and integrate this defense into our simulation environment, demonstrating its effectiveness under diverse conditions.

A former ByteDance intern who was allegedly dismissed for professional misconduct, including sabotaging colleagues' work, was announced as a winner of one of the most prestigious annual awards for AI research this week. Keyu Tian, whose LinkedIn and Google Scholar pages list him as a master's student in computer science at Peking University, is the first author of one of two papers chosen Tuesday for the main "Best Paper Award" at the Neural Information Processing Systems (NeurIPS) conference, the largest gathering of machine learning researchers in the world. The paper, titled "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction," presents a new method for creating AI-generated images that Tian and four coauthors--all affiliated with either ByteDance or Peking University--claim is faster and more efficient than its predecessors. "The overall quality of the paper presentation, experimental validation and insights (scaling laws) give compelling reasons to experiment with this model," the NeurIPS Best Paper Award committee wrote in a statement. The committee's decision to grant the honor to Tian, whom ByteDance reportedly sued for over 1 million in damages last month, claiming deliberate sabotage of other company research projects, quickly became the focus of wider discussions online about how NeurIPS is run and the way top AI researchers evaluate the work of their colleagues.