Apple has said it will update, rather than pause, a new artificial intelligence (AI) feature that has generated inaccurate news alerts on its latest iPhones.
“These AI summarisations by Apple do not reflect – and in some cases completely contradict – the original BBC content,” the BBC said on Monday.
A simple guide to help you understand AI Apple said its update would arrive “in the coming weeks”.
“A software update in the coming weeks will further clarify when the text being displayed is summarization provided by Apple Intelligence.
In November, a ProPublica journalist highlighted erroneous Apple AI summaries of alerts from the New York Times app suggesting it had reported that Israel’s Prime Minister Benjamin Netanyahu had been arrested.
Apple claims that a new artificial intelligence (AI) feature that has caused its most recent iPhones to send out erroneous news alerts will be updated rather than put on hold.
The company acknowledged the issues for the first time on Monday, stating that it was developing a software update to “further clarify” when the notifications are summaries produced by the Apple Intelligence system.
The technology’s poor performance has prompted calls for the tech giant to discontinue it.
Last month, the BBC lodged a complaint after an AI-generated synopsis of its headline misled some readers into believing that Luigi Mangione, the man suspected of killing Brian Thompson, the CEO of UnitedHealthcare, had shot himself.
Inaccurately summarizing BBC app notifications, Apple’s AI claimed that Rafael Nadal, the Spanish tennis star, had come out as gay and that Luke Littler had won the PDC World Darts Championship just hours before it started on Friday.
For the first time, Apple has formally addressed the BBC’s concerns regarding the errors, which seem to be originating from within the company’s app.
According to the BBC on Monday, “these AI summaries by Apple do not reflect – and in some cases completely contradict – the original BBC content.”.
As the accuracy of our news is crucial to preserving trust, Apple must take immediate action to resolve these problems. “,”.
How does artificial intelligence operate and what is it?
A basic guide to AI.
Apple promised that the update would be available “in the coming weeks.”.
Users can “scan for key details” with its notification summaries, which combine and rewrite previews of several recent app notifications into a single alert on their lock screens, according to a previous statement.
Receiving the summaries is optional, the company said in a statement on Monday, adding that “Apple Intelligence features are in beta and we are continuously making improvements with the help of user feedback.”.
“In the upcoming weeks, a software update will make it clearer when the text on screen is summarization from Apple Intelligence. Users who see an unexpected notification summary are urged to report the issue. “..”.
The feature was launched in the UK in December, along with other ones that were made available as part of its larger suite of AI tools. It can only be found on iPhone 16 models, iPhone 15 Pro, iPhone 15 Pro Max, and certain iPads and Macs running iOS 18.1 or later.
On social media, there have been multiple reports of the technology seemingly interpreting messages in a very literal and blunt manner.
A ProPublica journalist pointed out inaccurate Apple AI summaries of alerts from the New York Times app in November, implying that it had reported on the arrest of Israeli Prime Minister Benjamin Netanyahu.
The New York Times declined to comment, and the BBC has not been able to independently confirm the screenshots.
Reporters Without Borders, a group that advocates for journalists’ rights and interests, urged Apple to turn off the feature in December.
According to the statement, “generative AI services are still too immature to produce reliable information for the public” because the BBC was blamed for a false headline about Mr. Mangione.
Not only has Apple introduced generative AI tools that can produce text, images, and other content at user request, but the outcomes have varied.
Last year, Google’s AI overviews feature—which responds to user queries by providing a written synopsis of information from results at the top of its search engine—was criticized for offering inconsistent results.
A Google representative at the time claimed that these were “isolated examples” and that the feature was functioning properly overall.
