On July 1, Twitter users were blocked
from accessing the social media site and instead received an error message saying, “Sorry, you are rate limited. Please wait a few moments then try again.”
According to Twitter’s blog
, placing a reading limit allowed the Twitter teams to “detect and eliminate bots and other bad actors that are harming the platform.” Twitter advised that it gave no advance warning to its users to ensure that the bad actors would not have time to adapt their behavior and avoid detection.
While not explicitly referring to any of the target accounts, Twitter advised that the measures placed were to avoid bots from scraping public Twitter data to build and train AI models and to prevent accounts from manipulating users and conversations on the platform.
Originally, reading limits were set to 6,000 posts per day for verified accounts, 600 for unverified accounts, and 300 for new unverified accounts. As more users reacted to the unexpected limits, Musk tweeted that Twitter will increase reading limits without giving details on timelines or the new limits.
On Sunday, limits were effectively increased to allow verified users to read up to 10,000 posts per day, unverified users up to 1,000 posts per day, and 500 tweets for new, unverified users.
In addition to the read limits, Twitter also announced that users would need to log in to view posts
, a move Musk referred to as a “temporary emergency measure”. Twitter claims hundreds of organizations were scraping large amounts of Twitter data and the behaviors were impacting the Twitter user experience.
Musk has been vocal in his discontent with platforms using Twitter data to build and train large language models (LLMs), in particular calling out Microsoft and ChatGPT
. Musk has previously asked Microsoft to audit how it uses the social media data to ensure that it is complying with agreements that are in place. He has also threatened legal action.