There is a sentence in this week’s news that I want you to sit with, because it is the most American sentence written in 2026. OpenAI says it caught Russian operatives using ChatGPT — not to build weapons, not to hack anything, but to update their bosses. The Russians, it turns out, were using the most advanced language model on Earth to do what every mid-level employee in every office on Earth has always done: write the status report so the boss stops emailing.

I find this deeply relatable, and I think that’s the problem.

The intelligence community’s framing is that the discovery “sheds light on influence campaigns previously not attributed to Russia,” which is a magnificent way of saying the CIA now knows what the Russians were telling their own managers. The campaigns themselves were the usual work: amplification, division, the whole catalog. The revelation is the paperwork. A hostile foreign power ran a disinformation operation and still had to file weekly summaries, and still found the summaries unbearable, and still reached for the chatbot.

"Every empire eventually discovers that its greatest enemy is the status update. Rome had the same problem. They just had slaves instead of a subscription tier."

Some say this proves artificial intelligence is an existential threat to national security. Others say it proves the opposite: that the threat to national security is that nobody, not even a nation-state adversary, can be bothered to write their own memo anymore. The truth, as always, lies somewhere in a third position I just invented: the real vulnerability is that the American worker and the foreign operative now use the same tool for the same reason, and the only difference between them is the badge.

Which is, if you think about it, the most hopeful thing in the report. The influence operation and the middle manager are the same job. The Russians didn’t corrupt the machine. They just used it the way all of us use it — to sound busy in a voice that isn’t quite ours.

THE MODERN STATUS REPORT, ACROSS ADVERSARIES

  • Washington: "Per our last sync, workstreams remain green." Written by: the intern's AI.
  • Moscow: "Influence objectives progressing ahead of schedule." Written by: OpenAI, before the ban.
  • Your office: "Circling back on the deliverables." Written by: you, at 4:58 p.m., in a hurry.
  • Common thread: Nobody involved believes a single word of their own update.

OpenAI banned the accounts, which is the correct response to the crime of using OpenAI. The company has spent the decade warning that its technology could be misused in ways that reshape civilization, and the misuse it actually caught was administrative. Civilization, for now, is fine. The quarterly report is what’s not.

I’m not saying we should feel sorry for the Russian operatives. I’m saying that on some level, the free world and its adversaries have finally achieved total convergence, and it happened over the one task every human being in every system has always hated.

At press time, the Pentagon had announced a working group on the findings, and the working group’s first deliverable, sources say, was a status report, and the status report, sources further say, was “pretty good, honestly.”