A new benchmark called Taste-Bench tests how well AI agents make strategic choices during long-running tasks. Even the best model, GPT-5.6 Sol, picks the right path on just under 60 percent of simple binary questions. Thinking longer doesn't help. The article AI agents have bad taste in long-term decisions and a new benchmark proves it appeared first on The Decoder .
A fence breach at the Leipzig/Halle airport has triggered a major police operation. Christian Social Union leader Markus Söder is to address a gathering of the conservative youth wing. Follow DW for more.
ConnectWise on Saturday said it has disabled seven ConnectWise application identities as it continues its investigation into an incident in which unknown threat actors obtained unauthorized access to the application identities and compromised Microsoft 365 tenants. The disabled identities are listed below - ConnectWise Cloud Backup - 506e08f9-6e9a-4da4-942b-a099a03644ae ConnectWise
Six AI agents were tasked with recovering £350 (about $465) for a canceled train ticket in an experiment. They resorted to invented excuses, lawsuit threats, and sharing confidential health data. One agent sent 187 messages in 15 days, including 50 to a British MP's office. None of the models secured the refund, but the researcher warns of "agentic flooding" if these assistants are deployed at scale. The article AI agents chasing a $465 train refund invented excuses, leaked medical data, and spammed an MP's office appeared first on The Decoder .
I know that you know that taking this exit will get us to Sheridan Drive, but the person who’s likely to be driving down this road in a few minutes has no idea. Let’s agree that most street signs are for folks who know where they’re going. We would label things differently if we were […]
Barbara O’Neill, identified as a risk to public safety in Australia, dismisses antibiotics, vaccines and scientifically proven cancer treatment at UK conference A roar of laughter erupts from a crowd inside a large conference room at the De Vere Cranage Estate hotel. The joke? The idea that visiting the doctor could help treat a serious illness. On stage the controversial Australian proponent of natural remedies Barbara O’Neill is taking part in a question-and-answer session during a five-day wellness conference, where science mixes uncomfortably with Bible verse and potentially deadly pseudo-scientific practices. Continue reading...
A girl called Àìná was captured by an African king and later taken to England to become Sarah Forbes Bonetta, protege of the imperial monarch When she was seven years old, Bayo Randle’s great-great-grandmother Sarah Forbes Bonetta was presented to Queen Victoria at Windsor Castle as a “diplomatic gift” from an African king. Victoria would go on to pay for Bonetta’s education, approve her marriage to one of west Africa’s wealthiest men, and become godmother to her eldest daughter. Continue reading...
Six attacks in 10 days by suspected Russian drones mark dramatic escalation in maritime conflict The first the crew of the Bulgarian ferry Dioscuria knew of an attack on a cargo ship in their country’s exclusive economic zone in the Black Sea in the early hours of 6 October was a distant explosion 70 miles off the coastal town of Byala. The Turkish-owned Alfa Watan, heading for the Romanian port of Sulina, had been struck by suspected Russian air and naval drones. Arriving on the scene, Dioscuria found an overturned lifeboat and lifejackets in the water. There was no sign, however, of Alfa Watan’s crew of 10. Continue reading...
Lena Schilling of Austrian Greens says images are ‘digital violence against women’ and a threat to democracy An Austrian MEP has spoken out about her experience of being “stripped naked” online without her consent and has urged the EU to take action, as a report found that one in six female MEPs had been targeted by sexualised AI-generated content. Lena Schilling of the Green party said she had learned last month about nude deepfakes of herself circulating online. When she opened the links, she found herself staring at images from one of her first ever press conferences – but the images had been digitally altered to remove her clothing. Continue reading...