Google is introducing Agent Compute Engine, a runtime built specifically for AI agents that isolates each one in its own micro virtual machine. It is also broadening governance tools so companies can manage agents across cloud, hybrid and on-premises setups.
In a Latent Space interview, Modal CEO Erik Bernhardsson discusses elastic GPUs, the growth of sandboxes and the economics of cloud infrastructure for AI agents.
Sources at several US banks say AI agents are already getting into customer accounts and sometimes act without explicit user instructions. The reports raise questions about how banks will police automated access to sensitive financial data.
Anthropic is handing its Agent Skills standard to the Agentic AI Foundation, the Linux Foundation body it co-founded with Block and OpenAI and that already hosts MCP. Simon Willison notes he predicted the move when Anthropic first called skills an open standard.
A Verge column examines Mark Zuckerberg's vision of a network of autonomous AI agents that shop, work and talk to other people's agents on our behalf. It asks whether people actually want to communicate mainly through bot proxies.
The Nintendo Switch 2 appears to have been jailbroken, and a modder has already shown a custom version of Mario Kart World running on it. The hacker behind the "ns2jb" jailbreak says OpenAI's Codex helped build it, another sign that coding agents are being used for security research.