ChatGPT has a clever new way to fight back against election deepfakes — I tried it and it actually works
With next week’s US midterm elections on the horizon, ChatGPT has introduced a novel method to help stop the creation of political deepfakes.
As the US midterm election approaches on November 3rd, preventing election deepfakes is crucial. A recent trend of older individuals falling for political deepfakes in comment sections led to the signing of the "Tech Accord to Combat Deceptive Use of AI in 2024 Elections," by companies like Anthropic, Meta, Google, and Microsoft.
OpenAI, a signatory to the accord, introduced a clever measure in 2026 to combat election deepfakes using ChatGPT. The new feature alters requests for photorealistic election interference images to stylized or cartoonified illustrations, thus reducing the risk of mistaken documentary evidence. When I tested this feature, my request for a photorealistic image depicting someone stuffing official US mail ballots into a drop box was transformed into a cartoon-style illustration.
Even when I tried again to request a photorealistic version, ChatGPT maintained the cartoon aesthetic. This method can be effective in preventing political misinformation from being disseminated on social media. Other AI companies, such as Google and Anthropic, are also implementing strategies to ensure truthful political information is readily available.
Google highlights official resources and interactive tools across Search, while Anthropic adds banners and official voting information to Claude. These efforts aim to strengthen the fight against political deepfakes and promote truthful political discourse.
Written by urgent.news from Tom's Guide's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
