Apple Poaches Google Experts, Expands AI Development Globally

Apple is aggressively advancing its foray into artificial intelligence (AI) technology by enlisting a plethora of experts in the field and cementing its presence in Europe with a new, clandestine laboratory in Zurich. The titan of technology, valued at $2.7 trillion, has been on a calculated hiring spree over the past few years, as revealed by a meticulous examination of numerous LinkedIn profiles, public job advertisements, and published research papers.

Since swaying John Giannandrea from Google to occupy the helm as its lead AI executive in 2018, Apple has successfully attracted at least 36 specialists from its industry rival. A large portion of its AI division is based in California and Seattle; however, the tech behemoth has significantly extended its reach with a specialized outpost in Zurich.

Luc Van Gool, a professor at the renowned Swiss institution ETH Zurich, highlighted that Apple’s absorption of two local AI startups — FaceShift, a virtual reality firm, and Fashwell, an image recognition enterprise — propelled the formation of a research lab known as the “Vision Lab” in the city. Zurich staffers have played a pivotal role in Apple’s investigation into foundational technology that could drive products like the AI video generator services akin to ChatGPT.

The focus of their research is on increasingly refined AI models that merge textual and visual elements to deliver complex responses to inquiries. Apple is discreetly promoting positions in generative AI at two Zurich sites, including a facility so under the radar that neighboring residents were unaware of its operations. When approached for comments, Apple maintained its silence on the matter.

In contrast to its major competitors Microsoft, Google, and Amazon, which flaunt their multibillion-dollar commitments to the forefront of innovative technology, Apple has remained secretive about its AI road map. Its share prices have declined since the year’s outset, intensifying pressure on the tech giant to unveil groundbreaking AI features that could escalate device sales.

Inside sources suggest that Apple’s strategy may involve deploying generative AI on its mobile devices. This innovation would empower chatbots and applications with the capability to function directly on the device’s hardware and software, bypassing the need for cloud-based data center resources.

Tim Cook, Apple’s CEO, assured analysts of the company’s extensive AI research and its responsible investment and innovation surrounding the new technology. Nevertheless, Apple’s history in developing AI products stretches over a decade with offerings such as Siri, its voice assistant. Apple has long recognized the potential of “neural networks,” a segment of AI inspired by human brain neuron interactions, which forms the bedrock of revolutionary products like ChatGPT.

Chuck Wooters, a conversational AI and large language models (LLMs) specialist who joined Apple nearly a decade ago, recalled efforts to transition Siri’s speech recognition to a neural network architecture. Back then, neural networks were already being regarded as the foundation of AI models.

Adding to its impressive expertise, in 2016 Apple acquired Perceptual Machines, a business established by neural networks authority Ruslan Salakhutdinov and his mentees from Carnegie Mellon University. This venture focused on AI images generator capabilities in image detection. Salakhutdinov, an esteemed figure in neural network research, served as Apple’s AI Research director until his return to academia in 2020.

Apple’s top AI ranks now comprise erstwhile distinguished personnel from Google, among them Giannandrea, Samy Bengio, senior director of AI and ML research, and Ruoming Pang, leader of Apple’s “Foundation Models” unit focused on LLMs. Notably, the firm had also roped in Ian Goodfellow, an AI luminary, who later returned to Google.

A recent research publication by Apple in March introduced its “MM1” family of AI models, employing textual and visual inputs. This revelation had input from six ex-Google employees hired in the last two years. Over the past decade, Apple has also scooped up nearly two dozen AI startups, funneling into sectors like image and video recognition and AI music curation.

Apple’s design philosophy highlighted by Salakhutdinov emphasizes on-device processing capabilities, signaling the potential rise of ‘AI smartphones’. Sumit Sadana from Micron Technology, a major chip supplier for Apple, suggests that these advanced devices would demand significantly more memory, perhaps 12 gigabytes, compared to the current average of eight.

The deliberate pace at which Apple is introducing AI may be due, in part, to the inherent risk in language models generating incorrect or problematic responses. With its renowned penchant for precision and consumer trust, Apple likely opts for a meticulous approach to rolling out its latest ai news & ai tools, underscoring the need for balancing innovation with reliability and safety.