An account spreading videos likely generated by artificial intelligence (AI) has raised concerns over the growing use of synthetic media in the dissemination of political content in the United Kingdom. Experts from Resemble AI, a company specializing in detecting deepfakes and synthetic media, analyzed the videos and concluded with 97% confidence that they were AI-generated, citing irregularities in audio quality and distorted facial features.

Zohaib Ahmed, founder of Resemble AI, likened the phenomenon to social engineering attacks originating from foreign sources that target UK users with misleading information. Facebook is reportedly investigating the account following its identification but has not provided details on the timeline or potential corrective actions. The social media platform declined to confirm whether the account was monetized.

Academic experts emphasize the rising threat posed by AI-generated content. Dr. Lukasz Olejnik, visiting senior research fellow at King’s College London’s Department of War Studies, pointed out that AI-generated videos can create convincing but false visual evidence that is easily distributed, emotionally impactful, and challenging to verify promptly. He underscored the risk to UK democratic processes, noting that repeated exposure to such content could erode public trust in institutions, journalism, elections, and policy decisions, especially in sensitive political contexts such as security, migration, and foreign relations.

The concerns are not hypothetical. A 2024 study submitted to the House of Commons Home Affairs Committee linked AI-generated images depicting Muslims in threatening stereotypes with the Southport riots. Research from the London School of Economics (LSE) highlighted that social media posts containing racist conspiracy visuals were amplified 30% more than other content, fueled largely by far-right accounts. LSE warned that AI-generated materials, combined with algorithms that promote divisive content, contribute to accelerating radicalization and increasing the visibility of extremist viewpoints.

Against this backdrop, London Mayor Sir Sadiq Khan announced in June a £7 million initiative to counter online disinformation in the capital. The move followed a Greater London Authority report revealing a 200% surge over two years in posts framing London as declining and dangerous. Khan described the city as enduring a “relentless and unprecedented attack of lies and hatred.”

Ahmed stressed the scale of the challenge, noting that the creation and distribution of AI-generated content across social media channels can rapidly reach millions of users, making the phenomenon difficult to contain. Earlier this year, Meta’s independent Oversight Board urged the company to more proactively label AI-generated content and enhance detection capabilities.

The UK government’s Department for Science, Innovation and Technology reaffirmed that online platforms have legal responsibilities to address illegal content, including incitement to violence or hatred, and expect them to enforce strict measures. Facebook maintains guidelines prohibiting hate speech, misinformation, deceptive behavior, and coordinated inauthentic activity.

In a related development, Facebook removed several Vietnam-based pages identified as “content farms” after an investigation found they were spreading manipulated deepfake videos promoting false narratives about British politics. Efforts to contact the owners of the “Life in Britain” Facebook page linked to the posts were unsuccessful.