> he changed his mind not because of the pressure but a surge in credible violent threats stemming from the site.
How did he know they were credible and weren't just nonsense that was mass-produced by a GPT-3 clone?
It seems that nowadays, the requirements for kicking a site (with user-generated content) offline are: 1) a botnet with thousands of IP addresses, and 2) a decent GPU to automate coming up with posts that look like violent threats.
That sort of ruins Cloudflare's business model, but I half suspect that this new policy of the CEO won't be enforced consistently, and rather will depend on how many angry tweets are written. If that's the case, then the botnet and GPU from steps 1 and 2 can be used to create fake tweets too.
I'm not even saying he made the wrong decision, I'm just wondering what the next evolution of the meta-game is here.
How did he know they were credible and weren't just nonsense that was mass-produced by a GPT-3 clone?
It seems that nowadays, the requirements for kicking a site (with user-generated content) offline are: 1) a botnet with thousands of IP addresses, and 2) a decent GPU to automate coming up with posts that look like violent threats.
That sort of ruins Cloudflare's business model, but I half suspect that this new policy of the CEO won't be enforced consistently, and rather will depend on how many angry tweets are written. If that's the case, then the botnet and GPU from steps 1 and 2 can be used to create fake tweets too.
I'm not even saying he made the wrong decision, I'm just wondering what the next evolution of the meta-game is here.