Dignity vs. Deception: A Comparative Analysis of Non-Consensual Intimate Deepfake Regulation in the United Kingdom and the United States

The regulation of non-consensual intimate deepfakes reveals a transatlantic schism between protecting the dignity of the person depicted and protecting the liberty of the speaker. While the US Take It Down Act remains tethered to analogue-era concepts of specific intent and visual deception, the United Kingdom, through the Online Safety Act 2023 and section 66B of the Sexual Offences Act 2003, has pivoted towards a rights-based conception of sexual autonomy. The paper compares the two frameworks along three axes.

It begins with mens rea, contrasting the United Kingdom's consent-based offence, under which the non-consensual sharing itself grounds liability, with the American requirement of an intent to cause harm, which opens a 'satire loophole'. It then examines the actus reus, arguing that the US insistence that a forgery be 'indistinguishable from an authentic visual depiction' treats the wrong as a deception of the viewer rather than a violation of the victim and leaves 'cheapfake' victims unprotected, before contrasting the Take It Down Act's forty-eight-hour removal mandate with the Online Safety Act's ex ante safety duties. A blind spot common to both jurisdictions is then identified: the 'tool gap', whereby liability attaches to the user who publishes while the developers of abuse-generating tools remain largely unregulated.

Effective protection requires a hybrid model that joins British substantive definitions to American procedural speed, reinforced by cryptographic provenance standards and by liability extending upstream to the developers of these tools.