Michelle ThompsonApple has announced that it will update, rather than halt, a new AI feature after it generated inaccurate news alerts on its latest iPhones. In response to concerns, the company stated on Monday that it is working on a software update to “further clarify” when notifications are summaries created by its Apple Intelligence system.
The company is facing pressure to withdraw the technology following reports of errors, such as an AI-generated summary by Apple that falsely claimed Luigi Mangione, accused of killing UnitedHealthcare CEO Brian Thompson, had shot himself. Another misstep included an alert wrongly stating that Luke Littler had won the PDC World Darts Championship before it had started, and that Rafael Nadal had come out as gay.
This is the first formal response from Apple regarding the BBC’s complaint about these inaccuracies, which appeared as if they were originating from within the BBC’s app. The BBC emphasized that the AI-generated summaries did not reflect, and in some cases contradicted, the original content.
Apple’s update, expected in the coming weeks, will clarify when the notifications are AI-generated summaries. The company noted that the feature, which groups and rewrites multiple app notifications into a single alert, is optional and still in beta. The summaries aim to help users “scan for key details.”
The feature, launched in December, is available on iPhone 16 models, iPhone 15 Pro and Pro Max with iOS 18.1 or later, and select iPads and Macs. The technology has drawn attention for some highly literal interpretations of messages, with instances going viral on social media.
This is not an isolated issue, as other AI tools, such as Google’s AI overview feature, have also faced criticism for generating inaccurate content.
Emily Martinez
