Apple Introduces Groundbreaking AI at WWDC24

“Apple Intelligence,” the brand’s response to the likes of Google Gemini and Microsoft Copilot, did not display many features not already present in those other systems, but it did excel in two major areas: integration and privacy.

The technology, revealed in a prerecorded presentation on Monday at Apple’s annual World Wide Developers Conference, mainly leverages a user’s personal data to execute its AI tasks and performs much of this locally on the devices where the information is stored, thereby ensuring privacy.

“There are already some very impressive chat tools out there that perform a wide array of functions using global knowledge, but these tools know very little about you or your needs,” Apple Vice President of Software Engineering Craig Federighi stated during the presentation. “With iOS 18, iPadOS 18, and macOS Sequoia, we are embarking on a new path to bring you intelligence that understands you.”

He noted that Apple Intelligence can be utilized to comprehend and generate language, as well as images, and proactively simplify interactions across devices and apps. It can manage tasks like prioritizing notifications and offering writing tools to revise, summarize, and proofread text.

“They’re deeply integrating AI into your personal information and enhancing that data using artificial intelligence to make your apps smarter, integrate features, and extend capabilities,” explained Tim Bajarin, president of Creative Strategies, a technology advisory firm based in San Jose, Calif.

“It’s a unique approach,” he said. “Apple is the only one who can do this because they control the hardware, software, and semiconductors in their ecosystem.”

A Personalized Kind of AI

Mark N. Vena, president and principal analyst at SmartTech Research in Las Vegas, added that Apple’s vision of Apple Intelligence is centered around a personalized experience.

“People have been focusing on ChatGPT and other AI applications for publicly available content,” he commented. “Apple’s perspective is that the real value of Apple Intelligence lies in how users will utilize it in a highly personalized and customized manner.”

“Apple Intelligence will present a variety of interesting applications that the average individual will be able to leverage and find valuable,” he said.

“They’ve thrown down the gauntlet,” he declared. “They’re asserting that to effectively harness AI for the general public, it must be personalized, customizable, packed with meaningful applications, and built with a privacy component at its core.”

Private Cloud Computing

The cornerstone of Apple’s personal intelligence system is on-device processing, Federighi elaborated. This enables Apple Intelligence to be aware of personal data without gathering it.

Not all processing can be carried out locally, however. Some tasks might need to be performed in the cloud. For such tasks, Apple has developed “Private Cloud Compute,” which enables Apple devices to connect to specialized servers operating on Apple silicon for enhanced processing.

“These Apple silicon servers offer the privacy and security of your iPhone from the silicon on up, leveraging the security attributes of the Swift programming language, and run software with transparency built-in,” Federighi elucidated.

“When you send a request, Apple Intelligence determines whether it can be processed on the device,” he continued. “If higher computational capacity is needed, it can utilize Private Cloud Compute and send only the data pertinent to your task to be processed on Apple silicon servers. Your data is never stored or accessible by Apple.”

“And, much like your iPhone, independent experts can inspect the code running on these servers to verify this privacy commitment,” he stated.

“If Private Cloud Compute stays true to its promise, it’s a commendable effort that differentiates Apple significantly,” Eric Abbruzzese, a research director at ABI Research, a technology advisory firm headquartered in Oyster Bay, N.Y., remarked.

Vena was impressed with Apple’s diligence in constructing the privacy foundation for Apple Intelligence during the presentation. “Considering all the usage models they demonstrated require access to emails and text messages, it was crucial to convince users that privacy is central to their AI approach,” he observed.

However, not all AI inquiries go to Apple servers. OpenAI’s ChatGPT is also part of the mix. Siri might determine that a query is better answered by ChatGPT rather than personal sources. In that scenario, Siri would request permission to send the query to the OpenAI chatbot.

“Apple managed to get OpenAI to agree not to log users’ requests, and no personal data can be shared with ChatGPT without explicit permission,” Bajarin explained. “It’s an excellent approach and extremely secure.”

AI on Apple’s Terms

Ross Rubin, the principal analyst with Reticle Research, a consumer technology advisory firm in New York City, highlighted that Apple Intelligence provides the company with an opportunity to leverage the power of its processors for more than just rendering video or improving gaming performance.

“It allows them to bring more of these AI models onto the device, enhancing their privacy stance,” he pointed out. “While Apple Intelligence engages with a lot of personal data, none of that information goes anywhere.”

“This is AI on Apple’s terms,” he declared.

“No one will claim that this is the first time AI has done this,” he added. “The novel aspect is how they’ve integrated it into their apps in a way that represents a more comprehensive approach than what we’ve seen to date.”

Anshel Sag, a senior analyst with Moor Insights & Strategy, a technology analyst and advisory firm based in Austin, Texas, concurred.

“They’re not doing anything particularly novel, but their platform-level integration is efficient and appealing,” he commented. “And they’re presenting a robust privacy and computing narrative, which aligns with the expectations from Apple.”

“What they announced is similar to what other companies are offering, but not on this scale,” he elaborated. “What you’re seeing with Apple Intelligence is a more refined, cleaner integration of what others have already implemented, with a slight Apple twist. It’s not particularly groundbreaking.”

Beyond a Simple Value-Add

Abbruzzese observed that Apple has approached AI similarly to how they have approached other products.

“They’re never the first to market. They aim to be the best,” he noted. “It’s too early to determine if they are the best, but the cohesiveness of the announcement struck me. Everything came together seamlessly.”

“I haven’t observed the same from other players,” he continued. “AI often felt like an additional feature to something else, not as well-integrated.”

“Every AI feature Apple introduced operates across their ecosystem,” he added. “We haven’t seen such extensive synergy before. Microsoft Copilot is powerful, but it doesn’t feel as seamlessly integrated into the Windows ecosystem as Apple Intelligence.”