System prompts have emerged as a critical control surface for specifying the behavior of LLMs in chat and agent settings. Developers depend on system prompts to specify important context, output format, personalities, guardrails, content policies, and safety countermeasures, all of which require models to robustly adhere to the system prompt, especially when facing conflicting or adversarial user inputs. In practice, models often forget to consider relevant guardrails or fail to resolve conflicting demands between the system and the user. In this work, we study various methods for improving system prompt robustness by creating realistic new evaluation and fine-tuning datasets based on prompts collected from from OpenAI's GPT Store and HuggingFace's HuggingChat. Our experiments assessing models with a panel of new and existing benchmarks show that performance can be considerably improved with realistic fine-tuning data, as well as inference-time interventions such as classifier-free guidance. Finally, we analyze the results of recently released reasoning models from OpenAI and DeepSeek, which show exciting but uneven improvements on the benchmarks we study. Overall, current techniques fall short of ensuring system prompt robustness and further study is warranted.

After months at sea, the U.S. Coast Guard Cutter Waesche returned to San Diego on Thursday, with over 37,000 pounds of confiscated cocaine and one less crew member, lost at sea, officials said. The offloading of their massive narcotics haul -- which weighs about as much as a full grown humpback whale and is estimated to be worth 275 million -- comes days after search efforts were ended for 23-year-old Seaman Bryan Lee, according to the Coast Guard. Lee, who hails from Rancho Cordova, was discovered missing at 6:45 a.m. last Tuesday while the Waesche was conducting a routine counter-drug patrol around 300 nautical miles south of Mexico. Search crews dedicated over 190 hours to scouring 19,000 nautical miles for Lee using drones, aircraft and vessels, before suspending the search on Monday. The confiscated cocaine was netted through 11 drug interdiction missions off the coasts of Mexico and Central and South America from December through mid February.

Kurt "The Cyberguy" Knutsson explains how facial recognition technology can help you find your perfect match. From a poisonous date to finding love with a serial killer, these six chilling cases show how unsuspecting dating app users on the quest for romance led them into the clutches of danger. Dating apps – from Tinder to Grindr – are the modern way for people to connect with potential partners from the comfort of their own space. Brace yourself for stories that blur the line between love and terror. Here is Fox News Digital's list of some recent cases where love went wrong.

Apple's artificial intelligence features for iPhones could be available in China as early as May, according to Bloomberg. The company reportedly established several teams in China and the US to make that happen, and it's also teaming up with local companies for its generative AI needs in the country. Joe Tsai, Alibaba Group's Chairman, recently confirmed that Apple will use his company's generative AI technology for Chinese iPhones during an event. Tsai didn't say when Apple intends to roll out the AI features that use Alibaba's tech, but The Information previously reported that the companies had already submitted them for approval to the country's regulators. Bloomberg says Apple will use Alibaba's technology for its on-device AI models, specifically as a layer on top that can censor certain materials and information for the Chinese government.

As the electric-vehicle war in China calms down, leaving a few established players to dominate the field, Chinese EV giants are expanding into humanoid robotics. The shift is driven by financial necessity, but also by the advantages these companies command in the new sector: strong existing supply chains and years of experience building cutting-edge tech. The Chinese government is starting to promote and subsidize the transition, too. It's becoming clear that China is now committed to becoming a global leader in robotics and automation, just as it did with EVs. Over the past couple of weeks, I've been speaking to people whose voices have been recreated with AI.

A Russian drone with a high-explosive warhead has hit the Chornobyl nuclear power plant in the Kyiv region, Ukraine said, amid warnings by the military that Russia launched 133 unmanned vehicles against the country. Ukrainian President Volodymyr Zelenskyy said on Friday that the drone strike significantly damaged the protective containment shelter and started a fire, which has been put out. The Kremlin responded saying Russia does not hit nuclear sites. Radiation levels at the site have not increased, according to Zelenskyy and the International Atomic Energy Agency (IAEA). The IAEA did not attribute blame but said the drone strike occurred at 01:50am local time (23:50 GMT) and that there was "no indication of a breach in the … inner containment" shell, a protective cover built around the fourth reactor of the plant.

US Air Force Gen. Gregory Guillot told a senate committee that last year there were 350 drone detections over 100 military installations last year. A senior U.S. Northern Command (NORTHCOM) official told members of the Senate that some of the 350 drones that flew over military installations and sensitive areas last year may have been conducting surveillance. U.S. Air Force Gen. Gregory Guillot, who is commander of NORTHCOM and North American Aerospace Defense Command (NORAD), was questioned about the drones during a Senate Armed Services Committee Budget hearing on Thursday. Drones were spotted flying all over the country last year, though most notably in New Jersey. They were also flying over military installations, including Joint Base Langley, Wright-Patterson Air Force Base and Vandenberg Space Force Base.

Ukrainian President Volodymyr Zelenskyy said Friday that a Russian drone had struck a cover built to contain radiation at the Chernobyl nuclear power plant, adding that "radiation levels have not increased." The Ukrainian air force said that Russia had launched more than 100 drones across the country overnight -- including attack drones -- targeting northern regions of the country where the Chernobyl power plant lies. "Last night, a Russian attack drone with a high-explosive warhead struck the cover protecting the world from radiation at the destroyed 4th power unit of the Chernobyl Nuclear Power Plant," Zelenskyy said in a social media post. The International Atomic Energy Agency also reported an "explosion" at the site, and said "radiation levels inside and outside remain normal and stable." The agency, which has had a team deployed on the site since the early stages of Russia's invasion of Ukraine, published images apparently showing the drone on fire after crashing into the covering.

As Vice President JD Vance leaves Paris after urging Europe to reduce regulations and promote AI innovation, that effort is already in jeopardy. A series of quiet maneuvers by the Biden administration, major technology incumbents and a government-funded nonprofit called the Future of Privacy Forum (FPF) threatens to impose sweeping AI regulations in American states -- even after President Donald Trump revoked the previous administration's restrictive framework. On January 23, Trump signed Executive Order 14179, "Removing Barriers to American Leadership in Artificial Intelligence," replacing Biden's command-and-control approach with a pro-innovation mandate to defend U.S. AI leadership against rivals like China. But remnants of the old policy survive in nonprofits like FPF, which are busy drafting state bills mirroring President Joe Biden's agenda. Public records confirm FPF was obligated nearly 5 million from federal agencies across FY24 and 25 under Biden.

A speech by the US vice-president, JD Vance, symbolised a fracturing consensus on how to approach AI. He attended the summit with other global leaders, including the Indian prime minister, Narendra Modi, the Canadian PM, Justin Trudeau, and the head of the European Commission, Ursula von der Leyen. In his speech in the Grand Palais, Vance made it clear the US was not going to be held back from developing the tech by global regulation or an excessive focus on safety. "We need international regulatory regimes that foster the creation of AI technology rather than strangle it, and we need our European friends, in particular, to look to this new frontier with optimism rather than trepidation," he said. Speaking in front of the country's vice-premier, Zhang Guoqing, Vance warned his peers against cooperating with "authoritarian" regimes – in a clear reference to Beijing.