Apple to Update AI Feature After Criticism Over Inaccurate News Alerts
Apple has announced plans to update a new artificial intelligence (AI) feature following criticism over its inaccurate news summaries. The feature, which generates notification previews on its latest devices, has been called out for misrepresenting news content, including headlines from the BBC.
On Monday, the tech giant acknowledged the issue for the first time, stating it is working on a software update to better indicate when notifications are AI-generated summaries created by its Apple Intelligence system. This follows growing calls for the feature to be paused or retracted entirely due to its mistakes.
BBC Raises Concerns Over Misleading Alerts
Last month, the BBC filed a formal complaint after an AI-generated summary inaccurately suggested that Luigi Mangione, accused of murdering UnitedHealthcare CEO Brian Thompson, had shot himself. The error led to widespread confusion among readers who believed the notification had originated from the BBC’s app.
Further incidents have since emerged. On Friday, the AI summarized BBC notifications to falsely claim that Luke Littler had won the PDC World Darts Championship hours before the event began. Another error stated that tennis star Rafael Nadal had announced he was gay.
“These AI-generated summaries do not represent – and sometimes completely contradict – the original content published by the BBC,” the broadcaster said on Monday. It urged Apple to address the issues immediately, stressing that the accuracy of its reporting is crucial for maintaining public trust.
Apple’s Response and Planned Changes
Apple has stood by its decision to refine, rather than suspend, the feature. In a statement, the company reiterated that the AI-generated summaries, introduced as part of a broader suite of tools, are optional for users. It also emphasized that these features are still in beta testing.
“A software update in the coming weeks will make it clearer when text displayed on notifications is a summary provided by Apple Intelligence,” Apple stated. “We encourage users to report any concerns if they encounter unexpected or misleading notifications.”
The feature, available on iPhone 16 models and select iPads and Macs, compiles and rewrites multiple recent app notifications into a single alert, allowing users to quickly scan for key details. It was rolled out in the UK in December but has since faced backlash for its inaccuracies.
Growing Scrutiny Over Generative AI
Apple is not the only tech company navigating the challenges of generative AI. Similar issues have plagued Google’s AI-powered search summaries, which aim to provide concise overviews of information. These, too, have faced criticism for generating inaccurate or misleading responses.
Reporters Without Borders has called on Apple to disable the feature, citing concerns that generative AI remains too immature to reliably produce public-facing content. The organization pointed to the BBC incident involving Luigi Mangione as evidence of the potential harm inaccurate summaries can cause.
While Apple insists the technology is improving with user feedback, the errors have sparked a broader debate about the readiness of generative AI for real-world applications. For now, the company’s promised software update is set to roll out in the coming weeks, with hopes that it can restore trust in its AI systems.
