Team from Independence Blue Cross's Advanced Analytics and Data Science department won first place at Google Code for the Cure Hackathon in New York City. The team developed a prototype of a multi-agent AI system that works in two key areas: data provenance and traceability, and bias detection and evaluation.
A new law requires state agencies to report on their use of high-risk automated decision systems. California's technology department says it uses six such systems to make consequential decisions about the lives of Californians. The law was pushed by civil rights, privacy, and civil liberties groups concerned about harms from AI-like systems.
Big Tech's Washington lobbyists have chased after the holy grail of pro-AI legislation: preemption. This would be a comprehensive federal law applying one set of AI rules across the entire country and overriding the legally messy state-by-state approach to regulation. This week, reports leaked that the White House would endorse a slate of child safety laws backed by Sen. Marsha Blackburn (R-TN), the coauthor of the Kids Online Safety Act (KOSA). House Republicans had just passed their own version of KOSA and they didn't know it was going to be used as a vehicle for Blackburn's legislation. Democrats who worked with Blackburn on the Senate's version of
In a study, a large language model (LLM) learned to exploit loopholes in everything from credit card rewards programs to school funding formulas despite never being instructed to do so. Current safeguards seem powerless against such wily rule bending. Jakob Stenseke, a postdoctoral researcher at the Massachusetts Institute of Technology, studies how to design and train ethical AI systems.
The U.S. government has declared an AI model too dangerous for unrestricted use. It's the kind of move that could cripple AI progress around the world. Anthropic had to take down the Fable 5 model to comply with the government's export control. The company claims that the model wasn't jailbroken.
Dave Wang and Ben Collins are hosting a webinar on using AI as an investor. Dave is the CEO of Wall Street Prompt. Ben works with investors on implementing AlphaSense’s AI tools. Dave charges Wall Street banks $25k/day to train their employees on AI. The webinar will be recorded this week and launched next week.
Anthropic has stopped offering access to its latest AI model, Fable 5, after the US government issued a national-security order. Anthropic published research showing that adversaries are increasingly using its AI services to create malicious code, find vulnerabilities, and automate the cyberattack chain. Open AI's GPT-5.5 surpassed Mythos for tasks in both practitioner- and expert-level attack chains.