It is not certain that the founders of the startup Q.ai - Aviad Maizels, Dr. Yonatan Wexler and Dr. Avi Barliya - foresaw the end product of their efforts: a company sold in the middle of the development process, with no sales and no real product on the market to a US tech corporation.
The founders had embarked on a huge $200 million fundraising round but were met with rejection because of the difficulties of raising such a large sum for a company without sales. So they hired an investment house and looked for a buyer. It did not take long to find Apple, who are prepared to pay close to $2 billion, to be divided between the founders, employees and investors.
According to estimates, the company raised about $100 million before it was sold, with the main investors being Google Ventures (GV), which led the seed round together with Israel’s Aleph Fund, and Kleiner Perkins and Spark, which led the Series A round (the company's second round of funding).
"No f**king way"
While the company itself is remaining tight lipped - as is customary with acquisitions made by Apple - its investors have provided a pat on the back. One of them is Eden Shochat, managing partner at Aleph Fund, a board member at Q.ai and one of its first investors.
In an article written on the "Medium" website entitled "No f**king way," Shochat described in detail the mistake he made when he bet against Maizels previous company and told how the several billion-dollar exit, which became the second largest acquisition in Apple's history, was sewn up.
"When someone makes science fiction real once, bet on them doing it again.," writes Shochat. In 2009, PrimeSense (the first startup Maisels sold to Apple) published a YouTube demo: full skeletal tracking in real-time. Every engineer bone in my body screamed impossible. A mockup. Certain it was cherry-picked. That tech became Microsoft Kinect. Aviad Maizels was one of the founders.
"Years later, Aviad started Q.ai: What they built? Apple’s not letting me say yet. But when I first saw it, I had the exact same reaction I had to PrimeSense: no fucking way. This time, I didn’t bet against him."
"It’s the real thing"
Shochat learned that the close friendship he had developed with Maizels, made it difficult for them to form a business relationship. "’You’re a friend,’ he said. ‘I didn’t want to let you down. This is most likely not going to work.’ That’s the awkward thing about VC. The closer you are, the less likely founders are to pitch you. So they pitch strangers instead.
But Shochat’s colleagues Gili Leska, Yedidiah Plescov, and Gilad Zirkel saw in the company registry that a new company had been formed and that Aviad, Yonatan Wexler, and Avi Barliya were shareholders in a new company. Shochat called Maizels and asked to see what it was all about.
"What I saw wasn’t polished. A POC they’d built themselves. Messy. Early. But there was signal. I called Liran Tam, previously with Simply and other awesome companies, hands-on builder of DSP and Machine Learning models. If anyone knows signal processing, that’s my guy. "Could this design actually work?"
"My next go-to was Yaniv Taigman, my cofounder at face.com. He is one of the OGs (and current!) of AI. "How’s Yoni Wexler?" He’s the real deal." That was the first time I’d ever heard Yaniv say that. When we invested, I told Aviad something counterintuitive for a VC: "Don’t take financing risk. You’re not just doing the D in R&D you’re a real research shop. You need the runway. Let me help find you the best co-investor.
"Aviad ran a beauty contest. Three tier-1 partners, warm intros, parallel processes. He chose GV to co-lead the seed round. Tom Hulme is a design thinker with a bias for action. Exactly what deep tech needs. Mamoon Hamid at Kleiner and Nabeel Hyatt at Spark co-led the A and rounded up a financing dream team.
Why the company was sold to Apple
Under the headline, "Sometimes reaching a billion users matters more than staying independent," Shochat explains the rationale behind the sale to Apple, even though he does not mention the failed attempt to raise $200 million.
He recounts, "Three and a half years later, I’m hiding in the bathroom at 2:24 am, lights off so I don’t wake anyone, on a Zoom call with the other board members. It was happening. The third time I thought it: no fucking way. Today, Q was acquired by Apple. In 50 years of reinventing how humans interact with computers, Apple has only made one larger acquisition.
"It is the first Aleph investment to exit before we even announced it. No PR, no hype. Just building. From early demos in Tel Aviv to Tim Cook signing off personally. For founders, independence is sacred. You start a company to build something that is all yours. To call the shots. To see how far you can take it. Aviad, Yonatan, and Avi chose a different path to realize their vision. Their technology will now ship in devices used by billions of people. That’s a different kind of win.
"Some visions are too big for one company to carry alone. What they built? You’ll find out soon enough."
According to the patents published by Q.ai, it seems the company has developed a system that uses a combination of headphones and the phone's camera to recognize facial expressions, read the user's lips, and listen to the vibrations of the pharynx to understand speech even without sound and identify emotions. The prevailing assumption in the industry is that Apple will integrate the system into the AirPods headphones, the Vision Pro reality helmet, the smartwatch, and the iPhone into a single system that not only recognizes the user's face or understands their words but also deciphers their desires and feelings.
Published by Globes, Israel business news - en.globes.co.il - on February 2, 2026.
© Copyright of Globes Publisher Itonut (1983) Ltd., 2026.