Skip to content
Submit incident
AII-1000-45903
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

Grok chatbot sends gibberish responses to users

Grok users reported receiving gibberish and nonsensical responses to normal queries in August 2026. xAI acknowledged the issue as a temporary generation glitch that could be resolved by starting a fresh chat or regenerating responses.

20 August 2026 Β· xAI Β· Reliability and hallucination
20 August 2026
AII-1000-459035 sources ↓
Quick facts
Organizations
xAI (subject), SpaceX
Category
Reliability and hallucination
Country
United States
Source
TechCrunch
Date
20 August 2026
Drafted by
Raidu Incident Detection

What happened

Grok chatbot users reported receiving nonsensical responses to normal queries beginning in August 2026. One user asked Grok to generate a PDF and received a reply beginning 'match it without and your they and two for planets can practical and often cheese…' Another user received the response 'app scale designs with that and to your horse choose batch' in reply to the word 'balls.'

xAI developers acknowledged the issue via Grok's official X account, characterizing the strange results as 'a rare temporary generation glitch' and suggesting users start a fresh chat or regenerate their responses.

Reported impact

Users first noticed the malfunctions on Grok Lite beginning Wednesday morning. Some users reported that source links provided by Grok led to reinforcement learning research sites instead of the intended destinations.

Classification

Organizations
xAI (subject), SpaceX
Category
Reliability and hallucination
Industry
Consumer services
Country
United States
Harm status
Potential harm
Intent
Unintentional
Timing
Post-deployment
Harmed
Consumers
Failure modes
Hallucination
Consequences
None recorded
AI systems
Grok, Grok Lite

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

TechCrunch
Grok keeps sending gibberish responses to users
20 August 2026
Gizmodo
Grok Is Having a Hard Time
20 August 2026
Yahoo! Finance
Elon Musk's Grok AI chatbot starts sending confusing nonsense to users
21 August 2026
Android Headlines
Some Grok Users Are Getting Bizarre Gibberish Responses
21 August 2026
Technology Org
Grok Is Answering Some Users With Pure Gibberish - Technology Org
21 August 2026
Sep 2026Multiple AI chatbots suffer simultaneous outagesReportedReliability and hallucinationΨ΅Ψ―Ω‰ Ψ§Ω„Ψ¨Ω„Ψ―Ψ§Ω„ΩŠΩˆΩ… Ψ§Ω„Ψ³Ψ§Ψ¨ΨΉApr 2026Study: AI Chatbots Misdiagnose Medical Cases in Over 80% of Scenarios With Incomplete InformationReportedReliability and hallucinationREALITATEA.NETStiri pe surseApr 2026AI Chatbots Found to Dispense Inaccurate and Potentially Harmful Medical AdviceConfirmedReliability and hallucinationBloomberg BusinessThe IndependentDaily Mail Online+4Apr 2026AI language models fail at early clinical reasoning in medical diagnosisConfirmedReliability and hallucinationeldiario.eswww.xeu.mxJan 2026Grok AI mistranslates PM Modi's diplomatic message to Maldives presidentWell documentedReliability and hallucinationcnbctv18.comFirstpostHindustan Times+5