I’ve been trying for a long time to spend much less time on X. But in an attempt to not to dwell on the, well, absolute state of everything, I’ll try to say something nice: I didn’t entirely dislike the implementation of community notes. As rapidly developing stories pop off, it doesn’t hurt to have a little bit of extra context right next to the original post.
However, now I may have yet one more reason to leave for good, as X is piloting a scheme that allows AI chatbots to generate community notes. From July 1, X users can sign up for access to the AI Note Writer API in order to put forward their own little chatbot that can propose community notes. Like those written by humans, these AI-generated community notes will start appearing later this month on your feeds “if found helpful by people from different perspectives.”
This pilot scheme is being implemented in the hopes of both “[accelerating] the speed and scale of Community Notes,” and also leveraging feedback from the community in order to “develop AI agents that deliver increasingly accurate, less biased, and broadly helpful information.”
It’s worth noting this push comes after a reported 50% drop in the number of community notes created since January. Speaking to Adweek, an X spokesperson attributed at least part of this sharp decline to the passing “seasonality of controversial topics” such as the U.S. election. And as anyone who has spent time on X recently will have observed, there’s nothing for engagement quite like ‘controversial topics’.
In light of this AI note writer announcement though, my most pressing question is… did X forget the widely documented phenomenon of AI hallucinations? You know, that thing where LLMs tend to just make stuff up in a bid to only predict the word most likely to come next, rather than generating anything factual?
Introducing AI Note Writer API 🤖 AI helping humans. Humans still in charge.Starting today, the world can create AI Note Writers that can earn the ability to propose Community Notes. Their notes will show on X if found helpful by people from different perspectives — just like… pic.twitter.com/H4QNy6VTkwJuly 1, 2025
Okay, to be fair to X, it’s not going to be a complete AI free for all. For a start, AI notes “will be clearly marked for users,” and furthermore these will be “held to the same standard as human notes,” with “an open scoring algorithm to identify notes found helpful by people from different perspectives.” Still though, as with human-written community notes, I’d argue there’s still the possibility of disinformation gaining greater visibility through vote-bombing.
It’s all very ‘Move fast and break things’—and, speaking of, Meta notably let go of all of its third-party fact-checkers in favour of its own take on a community notes system earlier this year. Moderating massive platforms like Facebook and Instagram is a huge job, and one that is likely very expensive to even do somewhat well—offloading that work to a community of volunteers makes sense simply in terms of finances.
Though, that said, there’s likely other reasons why big tech like Meta is rallying around arguments of ‘free speech’ and shirking the responsibility of actually moderating their platforms. Memorably, Epic CEO Tim Sweeney recently criticised big tech leaders for “pretending to be Republicans, in hopes of currying favor with the new administration” in an alleged bid to skirt antitrust laws.
Former Microsoft head Bill Gates has attempted to offer another perspective, saying that though “you can be cynical,” he met with Trump in December “because he’s making decisions about global health and how we help poor countries, which is a big focus of mine now.” Besides all of that, Trump’s rolling back of Biden-era guardrails on AI and the ‘Big Beautiful Bill’ attempt to hold a moratorium on any state-level regulation of AI for the next decade suggests the current administration only intends to get cosier with big tech as the years drag on.
Best gaming PC 2025
