The issue primarily impacts users of the Grok Lite version, with complaints surfacing on Wednesday morning. While some individuals found that refreshing the session restored normal functionality, others noted that the gibberish responses persisted across multiple attempts. The bug appears confined to direct queries on Grok.com, leaving the chatbot's presence on the X platform unaffected.
Grok users report bouts of incoherent word salad
Users of xAI’s Grok chatbot are reporting a persistent technical glitch that causes the model to output strings of nonsensical text. Instead of coherent answers, the system has returned random word sequences and lists of research links, leaving the platform’s community forums flooded with confused reports.
In response to the mounting frustration, the official Grok account on X characterized the error as a rare, temporary generation glitch. Despite the user reports, the official xAI status page maintained that all services remained fully operational throughout the incident. The company did not provide further comment on the underlying cause of the failure, which follows a period of significant staff turnover that saw dozens of researchers and engineers depart the startup earlier this year.
Comments (0)
No comments yet. Be the first!