NCSC review issues AI warning ahead of UK election

Deepfake campaigns are likely to become more advanced in the run up to the next general election, with ‘high profile individuals’ becoming a focus for hackers, the UK’s National Cyber Security Centre has warned.

In its seventh annual review, the government body — which is part of intelligence unit GCHQ — also warned that nation’s critical infrastructure sectors also remained under ‘significant threat’.

The review highlighted a new trend of malicious actors targeting the personal email accounts of high-profile and influential individuals involved in politics.

Rather than a mass campaign against the public, the NCSC warns that there is a “persistent effort” by attackers to specifically target people who they think hold information of interest.

Earlier this year the cyber watchdog launched a new opt-in service for high-risk individuals to be alerted if malicious activity on personal devices or accounts is detected. The service also offers advise on further safety measures they can take.

Highlighting the threat posed by hackers using generative AI’s like ChatGPT and synthetic media (deepfakes) in the run up to the next UK general election — scheduled to take place in January, the watchdog stated:

“Large Language models (LLMs) will almost certainly be used to generate fabricated content; that hyper-realistic bots will make the spread of disinformation easier; and that deepfake campaigns are likely to become more advanced in the run up to the next nationwide vote, scheduled to take place by January 2025.”

The watchdog’s timely warning followed a deepfake audio of Sadiq Khan doing the rounds on social media this weekend that downplayed the importance of Remembrance Weekend commemorations.

There were concerns that the clip would further inflame tensions as hundreds of thousands of pro-Palestinian protesters gathered in London on Armistice Day.

While 145 protesters were arrested on the day, the London Mayor’s deepfake did not constitute a criminal offence, according to the Metropolitan police, which highlights how the law has not caught up with some of the new threats posed by AI.

At TechBBQ this year human rights activist Siyabulela Mandela, the great grandson of anti apartheid protester and South Africa’s first black president Nelson Mandela, warned how deepfakes could upend democracy and incite genocide in some unstable African countries.