As AI political deepfakes go viral, lawmakers struggle to keep up – 2 states’ laws show limits of what they can do

As AI-generated political deepfakes proliferate online, state lawmakers grapple with the challenges of regulating such deceptive content. Twenty-eight of the 31 states that have enacted election deepfake laws require disclosure of AI usage in campaign ads. In contrast, Maryland takes a more stringent stance, prohibiting certain deepfakes even when a warning is present.

For example, a fabricated video in Louisiana that manipulates a candidate's face or voice to falsely depict their statements can remain in circulation if it includes an AI warning. The warning, however, does not exempt the video from potential fines of up to $2,000 or imprisonment of up to two years if it damages a candidate's reputation or deceives voters.

On the other hand, Maryland's law prohibits certain deceptive deepfakes even when a warning is present. A violation can result in a fine of up to $5,000, up to five years in prison, or both. While a warning may suffice in Louisiana, Maryland's law requires actual or reckless creation, use, or spreading of a deepfake to be considered a violation.

The law also holds election officials responsible for correcting credible reports of misinformation about voting procedures, results, or rights, even if it means removing the content from circulation.

As the 2026 midterm elections approach, voters must consider the effectiveness of these laws in protecting them from AI-generated political deepfakes. Although the presence of a warning can satisfy disclosure requirements, it does not guarantee the accuracy of the content or prevent its dissemination. Maryland's stricter approach, which includes penalties for violators and the power to remove misinformation, offers a more comprehensive solution to combating the spread of false information through AI technology.