Kenya’s Foreign Affairs Principal Secretary (PS) Korir Sing’oei faced a backlash last Thursday after sharing an AI-generated deepfake on his official X account.
The video, designed to resemble a CNN broadcast featuring journalist Fareed Zakaria, falsely depicted Kenya’s role in Sudan’s peace efforts.
Users quickly flagged the video as fake, prompting the PS to delete it and issue a public apology. The incident ignited a national conversation on the risks of AI-generated content, with many warning of its potential for serious harm.
Deepfakes are digitally altered videos that make someone appear to be another person, often used to spread false information. Using machine learning, AI enhances lighting, speech synchronisation, and facial expressions, making detection difficult.
While distinguishing deepfakes from real content is increasingly challenging, social media analyst Egline Samoei suggests looking for unnatural facial movements, off lip-syncing, lighting inconsistencies, and using reverse video searches and fact-checking tools.
“There are some practical clues to look out for, such as watching for unnatural facial movements, off lip-syncing, lighting inconsistencies and using reverse video search and fact-checking tools. You can also check for traces of the same content online,” advises social media analyst Egline Samoei.
According to tech policy analyst Kennedy Kamande, an inadequate digital literacy across the population worsens the problem.
“Social media algorithms amplify sensational content, making misinformation spread faster than facts. The growing sophistication of AI-generated media makes it harder to tell real from fake,” he says.
“While deepfake technology was initially developed for entertainment, it has been weaponised for political manipulation, propaganda and gender-based attacks among other societal ills.”
The risks of deepfakes
Beyond embarrassment, sharing deepfakes can have serious consequences. “It fuels misinformation, and undermines the credibility of the institution or individual sharing it. In the end, people lose trust in them,” Ms Samoei explains.
“In some cases, it can even lead to defamation lawsuits or cyberbullying.”
Mr Kamande warns that deepfakes could be used to manipulate elections, influence policies, and shape public opinion.
“Globally, deepfakes have already been used for political deception. During the 2020 US elections, AI-generated videos spread false narratives about candidates. In India, deepfakes have manipulated voter sentiments. These cases highlight the dangers of unregulated deepfake technology,” he says.
Addressing the threat
Following the controversy, PS Sing’oei announced that the government is working with tech firms to explore watermarking AI-generated content.
“Additionally, we are designing – with the collaboration of partners and big-tech companies – the School on AI Diplomacy (SAID) at the Foreign Services Academy to equip policymakers better to navigate these emerging challenges,” said the PS.
Ms Samoei however observes that the watermarking initiative could turn out fruitless as there exists erasure mechanisms.
“The watermarks could certainly help in identifying AI-generated content, but it’s not a complete solution because those watermarks can be erased or cropped out,” she says.
Mr Kamande calls for an update of the country’s legal apparatus to criminalise deepfake abuse, in addition to upscaling media literacy amongst the population.
“The country must update its legal frameworks to criminalise deepfake abuse. Offenders should face strict penalties, and law enforcement agencies must be equipped with digital forensic tools to track and prosecute perpetrators,” he notes.
“Media literacy is another critical area that requires attention. Public education campaigns must teach Kenyans how to recognize deepfakes and critically assess digital content before amplifying it.”
