Meta’s account of the Claude AI Instagram suspension changed between Saturday and Sunday. On the day itself, Andy Stone, Meta’s vice president of communications, said very high rates of posting, sharing or commenting had triggered automated systems, and he called the action clearly a mistake.
Organizers pushed back on that version, saying they had posted only a few times and livestreamed to thousands of viewers. On Sunday, Stone posted a fuller explanation on X. He said Instagram’s engineers found that a Claude tool was reading profiles, posts and hashtags in short, intense bursts at about four times the usual volume. To Meta’s network security systems, that pattern looked like automated bot traffic rather than a person scrolling.
According to Stone, an automated re-check about an hour later recognized the removal was wrong and restored the account, but sorting out admin access took extra time. He repeated Meta’s apology for what he called the error.
What Meta did not say is just as important. Stone did not name the specific Claude tool, who was running it, or what it was doing with the No Kings account. The reports do not include any response from Anthropic, so it is still unclear whether the tool belonged to the coalition, a supporter, a researcher or someone else.