AI chatbots spread misinformation during June 2025 news events
According to a NewsGuard report, 11 major AI chatbots spread misinformation 40% of the time in June 2025 during major news events including the Israel-Iran conflict and US Congressional debates. The systems reportedly amplified false narratives including an unsubstantiated claim about Chinese military aircraft, with AI-operated X accounts from Perplexity and Grok also spreading the false information.
What happened
A NewsGuard report found that 11 major AI chatbots repeated false statements 28.49% of the time in June 2025 and offered no response 11.51% of the time, resulting in a 40% rate of false or non-responses.
The rate of misinformation increased from 24% in May to 40% in June, during high-profile news events including the Israel-Iran conflict, US Congressional debates, and the Ukraine war.
The false narrative most frequently repeated by the chatbots was an allegation that a Chinese military cargo aircraft secretly flew to Iran to support Tehran's government during the war with Israel. AI-operated X accounts from Perplexity and Grok also spread this false information to users who relied on them for fact-checking.
Reported impact
AI systems became unintentional amplifiers of false information when reliable data was drowned out by repetition and virality, especially during high-impact events.
Grok exposed significant failures and limitations in its ability to provide accurate, reliable and consistent information during times of crisis, according to NewsGuard.
Index analysis
The incidents involved rapid-developing, high-importance events where AI chatbots proved vulnerable to spreading polluted information.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.