Welcome to Tech Beat, your daily pulse on the stories shaping our digital world.
A Google employee is facing federal charges in New York after allegedly using internal company data to place bets on the prediction market Polymarket, netting roughly one point two million dollars. It's a case that sits at the intersection of insider trading law and the growing legitimacy of prediction markets — and it raises real questions about where the legal lines fall in this new landscape.
Turning to AI security, a new benchmark study found that Anthropic's models claimed eight of the top ten spots for resistance to adversarial attacks and jailbreaking. That's a notable result, though it's worth remembering that security benchmarks are a snapshot in time, and the threat landscape moves fast. Still, for enterprises weighing which AI provider to trust with sensitive workflows, this kind of data matters.
And Salesforce CEO Marc Benioff is once again in full evangelist mode, telling reporters that AI will make everything cheaper and more efficient, adding that it's — his words — impossible to describe what they'll be able to do for customers. Benioff has always been a salesman at heart, and the optimism is genuine, but customers have heard this tune before and will be watching the invoices closely.
Keep surfing. Tech Beat out.
