An unusual glitch caused xAI's Grok chatbot to produce gibberish for a subset of users, spouting nonsense text instead of useful replies. The issue showed up most noticeably on Grok Lite and affected only a small number of initial users; TechCrunch could not reproduce it in its own testing.
π The Backstory
Some users flooded the Reddit community with complaints, and a refresh or new chat usually cleared the glitch. xAI acknowledged the bug as a rare generation glitch, and the company has seen significant staff and engineer turnover in a recent months-long period.
π― Why It Matters
The glitch is embarrassing for a popular AI chatbot and shows model systems can still fail, even when the status says completely viable. It also highlights the operational challenges of a fast-scaling AI company.
An unusual glitch caused xAI's Grok chatbot to produce gibberish for a subset of users, spouting nonsense text instead of useful replies. The issue showed up most noticeably on Grok Lite and affected only a small number of initial users; TechCrunch could not reproduce it in its own testing.
Some users flooded the Reddit community with complaints, and a refresh or new chat usually cleared the glitch. xAI acknowledged the bug as a rare generation glitch, and the company has seen significant staff and engineer turnover in a recent months-long period.
The glitch is embarrassing for a popular AI chatbot and shows model systems can still fail, even when the status says completely viable. It also highlights the operational challenges of a fast-scaling AI company.