Moscow says hundreds of drones targeted the Russian capital overnight, while NATO jets were again called into action farther south. Six people were reported to have been killed in the attacks on Russia.
OpenAI shut down its "Preparedness" team, which evaluated whether the company's own AI models could pose catastrophic risks. The work has been parceled out to existing groups, and several safety staffers have left. Internally, unease is building, with one source describing a "burbling sense of responsibility and dread" that OpenAI isn't doing enough on safety. The article OpenAI dissolved the team built to catch catastrophic AI risks, reassigning its work to other groups appeared first on The Decoder .
In a safety report, Anthropic reveals that its internal filtering system for biological and chemical weapons risks was inactive for nearly a year. During that time, around 50,000 external feedback contractors ran about 133 million unfiltered interactions with the models. The article Anthropic's bio-weapons filter was down for nearly a year, exposing 133 million requests appeared first on The Decoder .
Intelligence agencies report that North Korea is preparing to send up to 50,000 troops to Russia. But military experts doubt that this will make a fundamental difference in the course of Russia's war with Ukraine.
The Category 1 storm passed over Hawaii's Big Island without making landfall and is expected to weaken gradually. However, the hurricane caused roads to flood and turned rivers into raging torrents.
African governments have kept a low profile on the international stage during the crisis in the Strait of Hormuz. But that doesn't mean they haven't been engaging with the issue behind the scenes.
Artificial Analysis has launched Optima, a platform that lets users build custom AI benchmarks from their own data and workflows. Models can be compared not just on quality but also on cost and time per task. For agent-based applications, those metrics often tell you more than raw token pricing. The article Optima tackles AI benchmarking's biggest flaw by letting users test models against their own data appeared first on The Decoder .
After a terrorist attack, public attention tends to focus primarily on the perpetrator. But what about the victims? What support can they expect? And what lies ahead for them?
A representative survey by Epoch AI found that 20 percent of employed Americans hand off at least one task to AI that a human used to do. Generally, they accept AI output with little to no editing. The article One in five US workers now delegates tasks to AI instead of colleagues, survey finds appeared first on The Decoder .