Apple has announced it will improve a controversial AI feature that has sparked confusion due to misleading news alerts on its latest iPhones. Following a complaint from the BBC regarding inaccurate summaries generated by its Apple Intelligence system, the tech giant stated it would implement software changes to clarify when notifications contain AI-generated content. The BBC raised alarms after a faulty summary suggested that Luigi Mangione, charged with the murder of UnitedHealthcare CEO Brian Thompson, had killed himself.
Moreover, a recent inaccurate notification claimed that Luke Littler had won the PDC World Darts Championship hours before its actual commencement and erroneously reported that Spanish tennis player Rafael Nadal had come out as gay. This incident represents Apple's first formal response to the errors that were made within the BBC app, with the broadcaster emphasizing that the AI-generated content does not accurately represent its original articles.
In a statement released, the BBC insisted, “It is critical that Apple urgently addresses these issues, as the accuracy of our news is essential in maintaining trust.” Apple revealed that the update would be rolled out in the forthcoming weeks. The company clarified that its notification summaries are designed to help users "scan for key details," and noted that these features are currently in beta, reflecting ongoing improvements based on user feedback.
These notification summaries, introduced in December for iPhone 16 models, iPhone 15 Pro, and select iPads and Macs, have been criticized for their blunt, literal interpretations, which have garnered viral attention on social media. Previous erroneous AI summaries from the Apple Intelligence system, such as misleading alerts from the New York Times app regarding Israeli Prime Minister Benjamin Netanyahu’s supposed arrest, have also come to light.
In response to the problematic AI behavior, international advocacy group Reporters Without Borders urged Apple to pause the feature in December, citing concerns about the unreliability of generative AI services in delivering accurate public information. Like Apple, other companies have also launched generative AI projects with uneven success rates; Google faced scrutiny for erratic summaries produced by its AI features last year, although they attributed the issues to isolated examples rather than systemic flaws.
