Use this for now. There are two remaining parts to the Iron Slopdome shipping in the next 2 weeks. pic.twitter.com/buGteBKXDo
— Nikita Bier (@nikitabier) May 14, 2026
X appears to be tightening its anti-spam systems again, but some legitimate users now say they are getting blocked from posting or replying altogether, with warnings that their activity looks “automated” or even “malicious.”
The full error reads:
This request looks like it might be automated. To protect our users from spam and other malicious activity, we can’t complete this action right now. Please try again later.
We’re seeing complaints about the issue pop up on Reddit and X itself. In one Reddit thread, the OP says X seems to have “changed the spam criteria somehow and it has run AMOK,” claiming long-time accounts are suddenly being told their posts look like bot activity.
Similarly, other users are also discussing the same issue on a couple of other Reddit posts. The poster in one of the threads said that this seems to be happening with accounts that are very active. They also said they “cannot imagine paying premium for this.”
Meanwhile, X’s Head of Product, Nikita Bier, has been talking up “Iron Slopdome” as a multi-part effort to crack down on AI-generated reply spam, telling one user there are “two remaining parts” still shipping over the next couple of weeks.
That said, this whole push to combat spam isn’t exactly new for the platform.
Back in February, its API started blocking automated spam replies, making it much harder for third-party tools to spray generic comments everywhere. In March, users reported a wave of bans for “inauthentic” behaviors, followed in April by creators losing thousands of followers in a large-scale bot purge.
Iron Slopdome looks like the next step in that same push. For people who really do run spammy AI farms, this is probably bad news in a way that X will happily take. For regular users suddenly branded as automated or malicious, though, it’s starting to look like a pretty bad deal.
It’s worth noting, however, that the error itself isn’t exactly new. We’ve seen discussions about it from over a decade ago. The only thing that’s changed is how it’s now impacting users on a wider scale than ever before.
Featured image generated with AI

