Meeting regulatory requirements and industry standards for AI systems
The 15 July framework binds data centres: where they sit, what power and water they draw, what they pay for the grid, plus training on Australian creative work. It puts no new obligation on anyone building with models. What already binds you is the law you have: consumer law, APRA, ASIC, the Privacy Act. Neither document says who gets to say no.
Claude Tag puts @Claude in Slack as its own per-channel service account, not on behalf of the person who asked. That breaks the delegation model everyone converged on, where the audit log can still name a human. Useful, but it revives the old service-account problems with an AI on top.
Agent identity makes more sense as four stacked layers: a tamper-proof token format, cryptographic proof of which workload is running, a delegation chain that keeps the human as the subject, and a way to onboard to a service an agent has never met. Climb all four and you've proven who the agent is. You still haven't proven that what it did was okay.
The EU (9/10 risk), US (5/10), and Australia (6/10) take vastly different approaches to AI regulation. Build for EU standards globally: the Brussels Effect means you'll need them anyway.
AI agents are breaking traditional identity and access management systems. From impersonation risks to cross-domain delegation chains, enterprises need new frameworks that balance autonomous operation with accountability and security.
We translate academic benchmarks into actionable risk signals through our proprietary AI governance pipeline, helping organizations implement regulatory-ready controls.
