OpenAI restricts photorealistic political deepfakes in ChatGPT Images ahead of US midterms
OpenAI is blocking photorealistic political deepfakes in ChatGPT Images, effective October 20. When users request images depicting election interference or voter suppression, the system now generates cartoon-style illustrations instead of realistic photos. This change aims to reduce the risk of misleading voters before the November 3 US midterms, where AI-generated content has become a concern.
Key points
- ChatGPT Images now generates cartoon-style images for political deepfake requests, not photorealistic ones
- OpenAI’s rule applies to content about election interference or voter suppression ahead of November 3 midterms
- Google and Anthropic lack specific public safeguards for AI-generated election content, per company statements
The move follows broader industry efforts to curb election-related disinformation. Google’s Gemini team references a midterm election policy but does not detail specific safeguards. Anthropic’s Claude model prohibits election-related deception under existing policies. OpenAI’s head of elections, Josh Lawson, noted the trade-off: while better image models improve creator tools, they also make AI-generated content harder to detect. The company has not confirmed whether the cartoonification rule will persist beyond the election cycle.
Model page: ChatGPT Images 2.5 →
OpenAI Has a Surprising Defense Against Election Deepfakes
cnet.com · 26 October 2026
Loading the full article…
This text was published by cnet.com and written by Katelyn Chedraoui. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Policy & Regulation
All →- Apple adds SynthID watermark to iPhone AI images · 1 src
- RAF chief says AI drones will accompany Typhoon fighters by decade's end · 1 src
- Will AI expose the rickety infrastructure of the internet? · 1 src
- Johns Hopkins notifies 3,182 patients over unapproved AI data leak · 1 src
- Anthropic’s S-1 warns catastrophic AI harm may block insurance · 8 src
Comments
via GitHub Discussions