X says its Safety team has identified a bot farm of roughly 200,000 suspected inauthentic Chinese accounts operating on the platform. A subset of 200 accounts was specifically dedicated to posting content about AI data centres. According to X's Global Government Affairs team, those accounts were posting "in a manner that could manipulate a legitimate debate about American AI and energy policy."
The posts reportedly claimed that AI data centres are driving up household electricity prices and putting excessive strain on local power grids. Some accounts also shared AI-generated cartoons depicting data centre operators profiting while ordinary households absorb the cost. One image showed data centre developers holding bags of cash next to the caption, "Data centres get subsidised while we foot the bill!"
Popular Now: DLSS 5 has been modded into Skyrim, and it definitely transforms character facesX didn't disclose exactly how it linked the accounts to China, nor did it specify what the remaining roughly 199,800 accounts in the broader bot farm were posting about. The company said it has suspended accounts found to violate its authenticity policy, adding, "We remain committed to maintaining an open and authentic platform where people debate topics of public interest. We take seriously any attempts to undermine the integrity of the global town square."
The anti-AI sentiment comes amid an increasingly heated domestic debate over AI data centre expansion. Rising electricity costs near data centres are well documented. Bloomberg has reported price increases of up to 267% in some areas close to major facilities, while Goldman Sachs found that electricity prices overall rose 6.9% year-over-year as of February 2026. Further increases are expected as more AI infrastructure comes online.
Frequently Asked Questions
Open a question for an answer from TweakTown's coverage of this news, or ask your own below.
How many accounts did X identify in the bot farm and how many focused on AI data centre content?
Did X explain how it attributed the accounts to China or provide technical attribution details?
What action did X take against the accounts that violated its authenticity policy?
Are there documented instances or examples of the AI-generated cartoons the bot accounts shared?
Have a question about this content not listed here? Ask below and TweakBot will answer it.
The incident also seems strangely similar to a case OpenAI reported in June. The company found a cluster of ChatGPT accounts likely based in China that were generating similar anti-data-centre content, including some of the same illustrations referenced in X's findings. OpenAI said the accounts primarily used Simplified Chinese prompts, often requested English-language outputs, and relied on VPNs to get around restrictions on mainland Chinese access to ChatGPT.
