When Apple Chief Operating Officer Tim Cook and his top lieutenants take the stage at the Worldwide Developers’ Conference in Cupertino on June 10, one announcement that may signal a new phase of the AI wars is the tech giant’s launch of generative AI across iPhone-specific applications like Siri and iMessage. However, the true star of the developer conference will be how Apple plans to roll out its AI features to its massive installed base of 2.2 billion iOS devices. Emphasis on building trust with consumers in the use of AI could be the differentiating factor that Apple might leverage to avoid the pitfalls that Google and others have encountered by deploying AI haphazardly in their messaging.
“It is understandable that these major corporations are in desperate competition for leadership, in one form or another, in developing and deploying AI,” Benjamin Kuipers, professor of Computer Science & Engineering at the University of Michigan, tells indianexpress.com. “Nonetheless, trust is an essential property for individuals, corporations, and society as a whole. When trust is lost, planning becomes more difficult, and carrying out those plans becomes both riskier and costly.”
In recent months, Google, OpenAI, and Microsoft have made headlines for all the wrong reasons, exposing another side of AI in contrast to the hype. First, Google had to roll back its AI Overviews search tool after the new feature produced inaccurate and odd results just two weeks after the announcement. Then, OpenAI came in for flak for introducing a voice for ChatGPT that sounded very much like Hollywood actress Scarlett Johansson, which the makers behind ChatGPT denied. However, this led to a larger debate on the new emerging threats in the AI age, such as who owns your voice. Microsoft’s AI ambitions were also questioned when a new AI-powered Recall feature that screenshots everything you do on your PC, as part of the new line of Copilot Plus PCs, was labeled a “disaster” for cybersecurity.
“The recent revolution in AI, building on deep neural networks, large language models, and generative AI, has enabled new levels of performance at the cost of incomprehensibility of the mechanisms and occasional significant hallucinatory errors,” Kuipers said, highlighting the recent case of Google’s AI Overviews search feature. He noted that these errors led to a huge loss of trust in Google.
The point that often gets missed in the larger AI landscape is the sense of urgency and the fear of not keeping up, making it hard not to join the AI movement. This gives big tech an upper hand, as whatever AI feature or tool they launch will attract users. However, the truth is that information on how the new technology works and ultimately benefits users and society at large is often not clear.
While many are questioning Apple for missing the bus and ignoring the potential of GenAI, the truth is that Cupertino has a history of being very cautious about jumping into new tech, and its stance on privacy is clear. According to experts, Apple’s AI strategy will likely center around keeping users’ information more private than how others handle it.
“Apple is not on the same AI cadence that Google is, nor is its entire business at stake,” noted Anisha Bhatia, senior technology analyst at GlobalData. “So Apple can – and will – approach generative AI cautiously.”
Bhatia agrees that Google doesn’t have the luxury to wait and watch before bringing AI to its core products and services, as the competition from OpenAI and Microsoft is cutthroat. However, according to Bhatia, Google does have a golden goose in the form of Android, the dominant mobile operating system that reaches billions of users. If Google deploys GenAI features faster, it only benefits the search engine giant by reaching consumers directly through products and services they are already familiar with.
Apple and Google may have different business models. However, both are platform holders with a massive installed base through their respective mobile operating systems and the grip they have through the applications that billions of people use daily. Google has already demonstrated what AI can do through its bespoke apps and products using an Android device. Apple, too, is charting a similar path by convincing the world how AI can become a bigger part of users’ lives. However, unlike Google, which works with different smartphone makers, each having a different timeline and flavour for how GenAI features come to Android smartphones, Apple will likely focus on AI tools that ordinary consumers can use in their daily lives and take advantage of, rather than fancy features that may not have as big an impact from a practical standpoint. Apple also has an advantage in capitalising on the giant user base that uses the iPhone, one of the most popular smartphones on the planet, which might play a big role in giving Cupertino an edge over others, even if the company may seem to be in catch-up mode with its competitors in AI.
Perhaps what will really be a conversation starter is how Apple solves AI’s biggest problems to its advantage by putting privacy at the center of everything. Apple doesn’t sell ads like Google and Meta, nor are its security claims dubious like Microsoft’s. Cupertino has leveraged privacy as a differentiating factor in the past. In fact, many users have bought Apple products because of the company’s privacy approaches. Experts say Apple will once again make privacy an issue—this time with artificial intelligence—and make a case around how user data security is under threat. Apple will demonstrate how GenAI features can now run on devices like iPhones and Macs natively, rather than in the cloud, thus maintaining control over user data and security, while also addressing response latency and the cost factor associated with running AI models in the cloud, and the energy required to operate data centers. For that, Apple will use its own large language models (LLMs) and won’t tap into user uploads to help train the AI, as many platform holders do. Apple will highlight how its hardware and software seamlessly integrate, and how it is utilising the “neural engine” as part of the chips in iPhone and Mac to run those AI models, which keep sensitive data only on their devices.
Insiders, however, warn that to scale AI more widely across personal devices, Apple may not only rely on supporting on-device capabilities. Eventually, it has to add hybrid AI capabilities. Partnering with an external resource like OpenAI and Google may be an option to bring cloud AI capabilities to the iPhone. But how it incorporates technology from an outside company that uses cloud processing while also protecting user privacy is something that needs to be addressed.
“The improvements in AI will be incremental, in measured doses, especially for consumer-facing tech, but there is no getting away from AI. Rest assured, the biggest news will be what AI features Apple brings to its devices, and how fast can it roll them out,” Bhatia said.
© IE Online Media Services Pvt Ltd
First uploaded on: 08-06-2024 at 08:25 IST
Deepak Sen is a tech enthusiast who covers the latest technological innovations, from AI to consumer gadgets. His articles provide readers with a glimpse into the ever-evolving world of technology.
