You're tuned in to Tech Beat. Here are the stories shaping the tech world right now.
OpenAI is making a notable change to how ChatGPT interacts with younger users, announcing the AI will no longer respond to teenagers in its human-like voice mode. The company says this is a proactive design choice, though it was careful to insist the move was not triggered by any specific incident involving children developing emotional attachments to the assistant.
Shifting to security, researchers have disclosed a prompt injection vulnerability in VirusTotal's Code Insights API, a tool powered by AI that helps analysts understand potentially malicious code. The finding is a reminder that when AI gets embedded into critical security infrastructure, it inherits an entirely new class of weaknesses that traditional software testing simply wasn't built to catch.
And in the world of crypto regulation, South Korea has become the latest in a growing list of more than thirty jurisdictions to restrict access to Polymarket, the prediction market platform. Regulators there rejected the company's argument that its peer-to-peer structure puts it outside gambling laws, ruling that managing market rules and facilitating crypto-based wagering is enough to bring it under their authority.
Keep surfing. Tech Beat out.
