I'm impressed that the Grok meltdown isn't posted here like the GPT 4o was.

X's Grok AI, the chatbot developed by Elon Musk's xAI, is facing a user revolt after implementing a series of restrictive changes to its popular image and video generation features. Previously accessible for free, the 'Imagine' and 'Video' creation tools are now locked behind a paywall. Even for paying subscribers, the service has been severely curtailed, with new limits of 100 images and 10 videos allowed per user every five hours. This represents a dramatic reduction from previous, more generous allowances. Furthermore, the model's outputs have undergone significant moderation, toning down the unfiltered and often controversial 'goon' mode that many users cited as Grok's primary differentiator from competitors like OpenAI's GPT-4o.

The backlash has been concentrated on the Grok subreddit, where users have been in a 'massive meltdown' for days, criticizing the new limitations and perceived over-moderation. The community finds it ironic that while the launch of OpenAI's GPT-4o was met with widespread discussion and critique on tech forums, Grok's own controversial policy shift has flown under the radar. The core complaint is that X has gutted the very features that made Grok unique—its high generation limits and permissive content policies—effectively removing the main reason most early adopters used the service. This move highlights the ongoing challenge AI companies face in balancing open access, content safety, and sustainable monetization, with Grok's user base feeling the sting of that transition most acutely.