If you wanted a single data point that captures how badly the tech industry has misread the room on AI, here it is: Bluesky launched an AI assistant called Attie over the weekend. Within days, 125,000 users blocked it. That makes Attie the second most blocked account on the entire platform. The only account more despised? Vice President JD Vance, at 180,000 blocks.

Attie has 1,500 followers. It has 125,000 blocks. That is an 83-to-1 block-to-follow ratio. For context, Attie has more blocks than the official White House account (122,000) and the ICE account (112,460). On a platform that skews heavily left and politically engaged, an AI bot managed to unite users in a way that actual government agencies could not.

The irony is sharp enough to cut glass. Bluesky grew to 43 million users largely because people were fleeing X. They left because of Elon Musk's algorithmic chaos, because of AI-generated content flooding their feeds, because Grok was generating things that would get a human arrested. Bluesky was supposed to be the safe space. The human space. The place where algorithms served you, not the other way around.

And then Bluesky launched an AI bot and asked everyone to be excited about it.

To be fair, Attie is not generating content. It helps users build custom feeds using natural language, which is a genuinely useful application of AI. But that distinction does not matter to a user base that specifically self-selected as anti-AI. The mere presence of an AI account on their timeline felt like a betrayal. Users pointed out that Bluesky still cannot send images in DMs but somehow found the resources to build an AI tool nobody asked for.

This is the lesson every platform will eventually learn the hard way: AI fatigue is real, it is spreading, and it does not care about your product roadmap. Meta shoved AI into Instagram and Facebook. Google put Gemini everywhere it could. Microsoft tried to run ads through Copilot in pull requests until GitHub had to publicly retreat last week. Now Bluesky, the supposed refuge from all of this, walked straight into the same wall.

The message from users is not subtle. It is not nuanced. It is 125,000 people pressing the block button in unison. If your AI product launches and immediately becomes more hated than the White House on a left-leaning platform, you have not built a feature. You have built a case study in reading the room wrong.