2h ago / 12:07 PM CDT
Share
AI services spread false information about shooting
Jared Perlo
For hours after the shooting, several generative AI platforms shared false, misleading or inaccurate statements in their responses.
Grok, the chatbot service from xAI integrated into X,
falsely labeled a
77-year-old former banker who was in Toronto as the suspect. Grok stated that The New York Times, CNN and Fox News had verified the allegation, but
none of the outlets had reported the man as a suspect.
Online misinformation watchdog
NewsGuard identified a
now-deletedpost from Perplexity AI's automated X account that stated Kirk was still alive more than 12 hours after he had been killed.
The
growing use of AI chatbots may help fuel the spread of misinformation online. A recent analysis found that generative AI tools were
almost twice as likely to spread false news information in August 2025 compared to the same time last year.
“So we must presume that the worst, rather than the best, choice will be made. The sober and responsible elements will be defeated in the present clash.” Philip K. Dick, The Man in the High Castle