It's not just new accounts. Since at least around mid-August they have been deactivating existing accounts (some 10+ years old) by forcing them into a broken "human verification" flow. You have to pass a Cloudflare check if using the web client, which then discards the signal and forces you to go through the app flow (which many report is also broken/stuck).

I suspect it's a crackdown on accounts used for services like xcancel/nitter, but their heuristics are laughably poor and have resulted in a massive wave of false positive effective "bans" while having no noticeable impact on mirroring services.

Much as I am loathe to excuse X for anything, I suspect that it’s due to the trend for governments putting age-related restrictions on exactly the sort of stuff that pervades X.

I’m going through the same thing setting up blewit [1] (a reaction to the traditional “sell the people’s data” social media sites) at the moment, had a phone call with the National Crime Agency, got all my Ofcom policies documented etc) - mainly because it’s a) user-to-user and b) because we will allow nsfw stuff after an age verification. That’s the law now in the UK and it has teeth - a criminal prosecution, not something a company can shield directors from.

1: https://blewit.net <- tap on ‘Open the sandbox’ for a demo

Why demand that users be human when Grok, the electronic "soul" of the site, isn't?

It's asymmetrical warfare against users.

Again.

Oh, and bot accounts aren't a problem. They're a feature. Grok can only lie, but it can't amplify the lies. Bot accounts do that, and do it well.

They cannot achieve their goal of becoming the US WeChat if users remain anonymous.