
Apple Inc. has delayed artificial intelligence (AI)-related updates for its Siri digital assistant, pushing back the heavily-hyped features at least several months in the latest hiccup at the company’s star-crossed AI division.
Several core features, including Siri’s ability to tap into a user’s personal information to answer queries and have more precise control over apps, were initially expected in the iOS 18.4 software update in April. Now, it may not be released until “the coming year,” according to a Bloomberg report.
Apple has been struggling to finish developing features as company engineers attempt unsuccessfully to fix bugs. What is more, fixing Siri will likely require more powerful AI models that run on Apple’s devices, straining hardware development, Bloomberg reported. That means Apple faces the grim prospect of either reducing its current feature set or make the AI models run more slowly on current or older devices.
Under a doomsday scenario, some within Apple’s AI division fear work on the features “could be scrapped altogether, and that Apple may have to rebuild the functions from scratch.” Such a delay would hold up a next-generation Siri until 2026, Bloomberg reports.
By any measure, the AI feature delays are a black eye for Apple, which has scuffled with AI at a time when its tech rivals — Amazon.com Inc., Meta Platforms Inc., Microsoft Corp. and Alphabet Inc.’s Google, to name a few — are making significant strides. Amazon is expected to debut its AI-turbocharged voice assistant Alexa+ this month. Google and Samsung Electronics Co. have also embedded AI into their devices.
All the more galling is that Apple has bomb-carpeted the airwaves the past six months with ads for the not-ready-for-prime-time AI features. When iPhone 16 was introduced last fall, the company told customers it was “built from the ground up for Apple Intelligence.” The truth now is that most of the core features might not be available until after iPhone 17 debuts later this year.
What is unclear is the potential impact the delay will have on Apple’s recent pledge to pour $500 billion into AI-related infrastructure spending the next four years which includes data centers and manufacturing facilities across the U.S.
AI preparedness increasingly has become a major concern throughout the industry. Last week, OpenAI’s former head of public policy accused the company of “rewriting” its AI safety and alignment history during the launch of Chat GPT-2 in early February.
“In a discontinuous world, practicing for the AGI moment is the only thing we can do, and safety lessons come from treating the systems of today with outsized caution relative to their apparent power,” Miles Brundage said in a blog post. “This is the approach we took for GPT‑2 when we initially didn’t release the model due to concerns about malicious applications.”