Apple has poached dozens of Google artificial intelligence experts and created a secret European lab in Zurich, as the tech giant builds a team to battle rivals to develop new AI models and products.
The company has been on a hiring spree in recent years to expand its global AI and machine learning team.
The iPhone maker has particularly attracted workers from Google, which has poached at least 36 experts from its rival since it made John Giannandrea its top AI executive in 2018.
While most of Apple’s AI team works out of offices in California and Seattle, the tech group has also grown a significant outpost in Zurich.
Luc von Goll, a professor at the Swiss university ETH Zurich, said Apple’s acquisition of two local AI startups — virtual reality group FaceShift and image recognition company Fashwell — has created a research laboratory in the city known as VisionLab.
Employees based in Zurich have been involved in Apple’s core technology research that powers products such as OpenAI’s ChatGPT chatbot. His papers have focused on more advanced AI models that incorporate text and visual input to generate answers to questions.
The company is advertising jobs in generative AI at two locations in Zurich, one of which is particularly low-profile. A neighbor told the FT that he was not even aware of the office’s existence. Apple did not respond to requests for comment.
Apple has generally been tight-lipped about its AI plans even as major tech rivals Microsoft, Google and Amazon have invested billions of dollars in the cutting-edge technology.
Its shares have fallen since the start of the year, while rivals’ stocks have risen, putting pressure on the tech giant to announce game-changing AI features that could boost device sales. .
Industry insiders suggest that Apple is focusing on deploying generative AI on its mobile devices, a development that would allow AI chatbots and apps to run on the phone itself rather than running through cloud services in data centers. Will allow running on hardware and software.
Chief Executive Tim Cook has told analysts that Apple is “researching a wide range of AI technologies” and investing and innovating in new technology.
However, the tech group has developed AI products for more than a decade, such as its voice assistant Siri. The company has long been aware of the potential of “neural networks” — a form of AI that uses neurons in the human brain to communicate and a technology that powers breakthrough products like ChatGPT.
Chuck Voters, an expert on conversational AI and big language models who joined Apple in December 2013 and worked on Siri for about two years, said: “While I was there, there was a push in the Siri group. was one. The move to neural architecture for speech recognition. Even then, before the big language models came along, he was a big proponent of neural networks.”
That interest seems to have led Apple to the researchers who were the driving force behind the neural networks that power AI models.
There was a company founded by Ruslan Salakhotdinov and two of his students at the university, which worked on creative AI-powered image detection.
“At the time they were hunting a lot of researchers and trying to build the infrastructure to train these models,” Saladinov told the FT.
Salakhutdinov is a key figure in the history of neural networks, and studied at the University of Toronto under the technology’s “godfather,” Jeffrey Hinton, who last year cited concerns about the dangers of creative AI. Google left. Salakhotdinov worked as director of AI research at Apple until 2020, when he returned to academia at Carnegie Mellon.
Apple’s top AI team is now made up of former Google executives, including Giannandrea, who previously oversaw Google Brain, the search company’s AI lab that has been merged with DeepMind.
Sammy Bengio, Senior Director of AI and ML Research, was previously one of Google’s top AI scientists. Roming Peng, who leads Apple’s “Foundation Models” team working on LLMs, previously led Google’s AI speech recognition research.
The company once hired Ian Goodfellow, another deep learning pioneer, but he returned to Google in 2022, protesting Apple’s back-to-work policy.
Six former Google employees over the past two years were listed among the authors of a major research paper published in March, in which Apple revealed that it had developed a family of AI models called “MM1.” which use text and visual input to generate responses. .
Apple has also bought nearly two dozen AI startups over the past 10 years, focused on applying AI reasoning to image and video recognition, data processing, search capabilities and music content production.
Among them, the founders of Musicmetric, Emotient, Silk Labs, PullString, CamerAI, Fashwell, Spectral Edge, Inductiv Inc, Vilynx, AI Music and WaveOne still work at Apple, according to their LinkedIn profiles.
Salakhutdinov said Apple has focused on “doing as much as you can,” which will require more powerful chips with so-called dynamic random access memory (DRAM) that can handle the large amounts of data needed for power. Can handle AI models.