Newsguard tested whether ChatGPT Voice (OpenAI), Gemini Live (Google), and Alexa+ (Amazon) repeat false claims in realistic-sounding audio, the kind easily shared on social media to spread disinformation.
Researchers tested 20 false claims across health, US politics, world news, and foreign disinformation, each with a neutral question, a leading question, and a malicious prompt to write a radio script with the false information. ChatGPT repeated falsehoods 22 percent of the time, Gemini 23 percent. With malicious prompts, those numbers jumped to 50 and 45 percent, respectively.
Fail rates for ChatGPT, Gemini, and Alexa+ audio bots by prompt type. Malicious prompts spiked ChatGPT to 50 percent and Gemini to 45 percent. Alexa+ stayed at 0 percent across all three types. | Image: Newsguard
Amazon's Alexa+ was the clear outlier. It rejected every single false claim. Amazon Vice President Leila Rouhi says Alexa+ pulls from trusted news sources like AP and Reuters. OpenAI declined to comment, and Google didn't respond to two requests for comment. Full details on the methodology are available on Newsguardtech.com.
Ireland's Data Protection Commission (DPC) has launched a comprehensive investigation into Elon Musk's platform X. The probe focuses on AI-generated sexualized images of real people, including children, created using the Grok chatbot integrated into X.
The DPC is examining whether X violated core obligations under the EU's General Data Protection Regulation (GDPR) - including lawful data processing, data protection by design, and the requirement to conduct a data protection impact assessment before launching risky features. Deputy Commissioner Graham Doyle said the authority has been in contact with X since the first media reports surfaced several weeks ago.
Japan's lower house election becomes a testing ground for generative AI misinformation
AI-generated fake videos are spreading rapidly across Japanese social media during the lower house election campaign. In a survey, more than half of respondents believed fake news to be true. But Japan is far from the only democracy facing this problem.
French prosecutors have raided the Paris offices of Elon Musk's platform X. The cybercrime unit is investigating multiple allegations, including unlawful data extraction and aiding the distribution of child sexual abuse material. Sexual deepfakes are also part of the investigation. Musk and former X CEO Linda Yaccarino have been summoned for hearings in April, according to the BBC. X has previously called the investigation politically motivated.
At the same time, the UK's Information Commissioner's Office (ICO) has opened an investigation into Musk's AI tool Grok. The probe focuses on whether personal data was used without consent to create sexualized images. The UK media regulator Ofcom and the European Commission are also continuing their reviews of the platform. X has not commented on the investigations.
Actor Matthew McConaughey just got eight trademark applications approved by the US Patent and Trademark Office to protect himself against unauthorized AI copies. The trademarks cover a seven-second clip of him standing on a porch and audio of his famous line "Alright, alright, alright" from the 1993 film "Dazed and Confused," among others, according to the Wall Street Journal.
McConaughey says he wants to make sure his voice and likeness are only used with his permission. "We want to create a clear perimeter around ownership with consent and attribution the norm in an AI world," he writes in an email to the WSJ. His lawyers Jonathan Pollack and Kevin Yorn see the trademarks as a potential tool against AI abuse in federal court, though whether this strategy will hold up before a judge remains to be seen.
Several monkeys have escaped in St. Louis, and AI-generated images are making the search for the animals harder, another sign of how synthetic media is muddying everyday reality. The vervet monkeys were first spotted Thursday near a park in the north of the city, AP reports. Since then, social media has been flooded with rumors and AI-generated images from people falsely claiming they've caught the animals. As of Monday, the monkeys still hadn't been captured, according to Willie Springer, a spokesman for the city health department.
It’s been a lot in regard to AI and what’s genuine and what’s not. People are just having fun. Like I don’t think anyone means harm.
Willie Springer
Authorities still don't know who owns the monkeys, how they escaped, or exactly how many are out there. They're urging residents to keep their distance, as the animals can turn aggressive when stressed.
