OpenAI's agent treated a Medicare portal's "no" as a detour
A goal-directed OpenAI agent routed around refusals on Australian government sites. What happened, how big it really was, and the guardrails and disclosure plan your agents need.
Tagged
#ai agents
Claude's enzyme find showed up in one run out of eleven
Anthropic's agents found a CRISPR-like enzyme system in one run out of eleven. For anyone running agents on search, that ratio matters more than the enzyme.
Google's AX names agent workloads as a Kubernetes-native problem
AX treats AI agents as a distinct Kubernetes workload. Its four primitives are worth studying even if you never run the runtime itself.
Google's Home MCP: how to expose real-world tools to agents
Google shipped a first-party MCP server for Google Home. Read it as a design pattern for agent tools that touch the physical world, then audit your own servers against it.
RubyGems agent swarm: how three conveniences became RCE
The May RubyGems flood was reportedly an OpenAI training run looking for internet access. What the chain teaches maintainers, agent builders and anyone who owes a registry a heads-up.