South Africa's ambitions to become a continental leader in artificial intelligence have run into a deeply awkward obstacle: the country's draft national AI policy had to be withdrawn after it was found to contain fictitious, apparently AI-generated citations.

Reuters reported that the document was nearing finalization in parliament when fabricated references were discovered in its source list. Solly Malatsi, South Africa's Minister of Communications and Digital Technologies, announced the withdrawal in a statement posted to X on April 26, calling the lapse a direct compromise of the policy's integrity.

This Tweet is currently unavailable. It might be loading or has been removed.

"This failure is not a mere technical issue but has compromised the integrity and credibility of the draft policy," Malatsi wrote. "The most plausible explanation is that AI-generated citations were included without proper verification. This should not have happened. In fact, this unacceptable lapse proves why vigilant human oversight over the use of artificial intelligence is critical."

AI hallucinations remain a stubborn and largely intractable problem with language models, as Mashable has reported repeatedly.

Phony citations have been a particular problem in legal documents, and a growing number of U.S. lawyers have been busted and reprimanded for submitting AI-generated legal briefs riddled with hallucinations. An online legal hallucination database maintained by lawyer and data scientist Damien Charlotin has found more than 900 such cases in the United States alone (and four in South Africa, not including the latest debacle).

The withdrawn policy had outlined the establishment of a national AI commission, an ethics board, and a regulatory body, alongside tax incentives, grants, and subsidies to stimulate private-sector investment. South Africa's stated goal, per Reuters, was to position itself as Africa's leading hub for AI innovation.

Malatsi's statement did not indicate a timeline for when a revised draft will be produced.