UK to force tech giants to remove deepfakes within 48 hours

# UK to force tech giants to remove deepfakes within 48 hours

The UK government has launched an initiative that could radically change how technology giants police content on their platforms. A bill introduced in Parliament sets an unprecedentedly tight deadline – just 48 hours – for removing unacceptable images created with artificial intelligence. The measure, designed to protect users from abuse and safeguard their private lives amid the rapid development of digital technology, responds to a recent incident that shook one of the largest social networks.

Context: The incident on X (formerly Twitter) exposed a weakness in modern platforms. Users were able to generate and spread deepfakes at scale, including explicit images, using the platform's own internal tools. Thousands of such AI-generated photos were published, prompting a wave of criticism and concern over the ethics of using these technologies and the responsibility of their developers. The case showed how easily privacy boundaries can be breached and people's reputations damaged when content-generation tools fall into the hands of bad actors or are used irresponsibly.

In-depth analysis: The new bill aims to place direct responsibility on technology companies for the safety of their algorithms and the content generated with them. The 48-hour limit for removing deepfakes, especially those that are abusive, defamatory or sexual in nature, is a strict requirement. It implies that platforms will have to do more than respond promptly to incoming complaints: they will also need to proactively detect and block such content.

That will require tech giants to invest in more advanced AI-based moderation systems, as well as in training staff who can quickly assess disputed material and make decisions on it. The bill also touches on algorithmic transparency and the potential influence of algorithms on the spread of disinformation and harmful content.

Consequences: Such tight deadlines pose serious challenges for technology companies. On the one hand, they may improve user safety and reduce cases of cyberbullying and the spread of unacceptable content. Users will feel better protected knowing that platforms are required to respond promptly to threats to their privacy. On the other hand, such short deadlines may lead to moderation errors, with legitimate content wrongly removed for fear of fines. There is also a risk that companies may begin to apply more aggressive censorship to avoid breaching the new rules, which may affect freedom of speech. Companies will likely have to revisit their policies on the use of AI tools and strengthen internal control procedures.

Conclusion: The UK bill is an important step in the attempt to regulate generative AI and protect users in the digital space. It reflects a growing understanding that technology companies bear responsibility for the consequences of using their products. The initiative's success will depend on how effectively platforms can adapt to the new requirements while balancing safety, freedom of expression and innovation. The international community will watch developments in the UK closely, since similar problems are relevant to every country actively adopting artificial intelligence technologies.