Hackers hit AI coding tool OpenCode, running up charges on users’ AI accounts
THE SO WHAT
Agentic dev tools that proxy user API keys are now a direct fraud surface—OpenCode’s breach shows the blast radius is not data loss but runaway model spend. If you ship AI tools that touch customer keys, move key storage, rotation, and spend caps into your core product requirements, not “nice-to-have” security hardening.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIPartnering with Blacksmith
If agents can write production-scale code, the new bottleneck is proving that code is safe, performant, and compliant at the same speed. Expect “agent-era CI/CD” vendors like Blacksmith to become a required layer in any serious agent deployment, not a nice-to-have devtool.
Applied AIAmazon is taking its biggest shot at Nvidia yet with new TorchNeuron dev tool
Amazon going after PyTorch lock‑in with TorchNeuron is a direct attempt to make switching off Nvidia GPUs feel incremental, not heroic. If you’re building new training workloads, you now have to evaluate AWS’s toolchain as a first‑class alternative rather than defaulting to CUDA by habit.
Fired OpenAI engineer accused of bugging office won't be released from jail
An alleged internal spying case at a major lab underlines that AI companies are now geopolitical intelligence targets, not just IP targets. Treat your own model, data, and leadership environments with the same physical and insider‑threat rigor you’d apply to critical infrastructure.
Applied AIOpenAI Engineer Accused of Spying on Executives Ordered to Remain in Custody
A judge calling a foreign government’s bail offer a “blank check” in an AI espionage case makes explicit how national interests now intersect with lab security. Boards and CISOs at AI‑heavy firms should be treating insider risk and foreign influence as board‑level agenda items, not HR issues.