Skip to main content
OpenAI logo over Apple Portrait Mode team members, symbolizing the $300M acquisition of their startup.

Editorial illustration for OpenAI Buys Apple Portrait Mode Team's Startup for USD 300M

OpenAI Acquires Apple Portrait Mode Team for $300M

4 min read

OpenAI has agreed to buy Glass Imaging, a small camera-technology startup in Los Altos, California, for more than $300 million, according to a Wall Street Journal report. The company was founded in 2019 by Ziv Attar and Tom Bishop, two former Apple engineers who once led the team behind Portrait Mode, and it had raised roughly $30 million from investors before this deal. OpenAI did not respond to a request for comment on the acquisition.

The purchase fits a pattern. OpenAI has been building out a hardware arm for months, with reports pointing to smartphones, earbuds, and AI companion gadgets in various stages of development. Sam Altman already made his hardware ambitions public earlier in 2025, when he and Apple's former design chief Jony Ive announced they'd secretly been running a device startup called io, which OpenAI then bought for $6.5 billion.

Glass Imaging's specialty is squeezing better photos out of the tiny lenses and sensors packed into phones, a problem Attar and Bishop spent years wrestling with at Apple. Their approach to that engineering challenge, and why it might matter to a company that just spent billions on hardware, is where the deal gets interesting.

Glass Imaging was founded by Ziv Attar and Tom Bishop, a pair of former Apple engineers who previously led the team that developed Apple’s Portrait Mode. This background directly informs the work they now do at Glass Imaging, where they use AI to overcome the physical size constraints of smartphone cameras.

Why this matters

OpenAI has now bought a camera hardware company, and that's worth sitting with for a second. This isn't a language model tweak or a chatbot feature. Attar and Bishop spent years solving a specific physical problem: how to get good depth and low-light performance out of a lens that's smaller than a coin.

That's optics and computational imaging, not next-token prediction. If OpenAI is folding that expertise into its stack, it points toward hardware that sees the world directly rather than through a phone's camera app, think custom devices with cameras tuned by people who built Portrait Mode. For founders, the signal is that foundation model companies are willing to spend $300 million on teams with narrow, physical-world skills, not just ML researchers.

For researchers, it's a reminder that multimodal AI's next bottleneck might be the sensor, not the model. Worth watching whether OpenAI ships a device, licenses the tech to partners, or just wants the talent. At $30 million raised against a $300 million exit, Glass Imaging's investors did fine either way.

Common Questions Answered

Who founded Glass Imaging and what was their background at Apple?

Glass Imaging was founded in 2019 by Ziv Attar and Tom Bishop, two former Apple engineers who previously led the team that developed Apple's Portrait Mode. Their expertise in computational imaging and smartphone camera technology directly informs the work they continue at Glass Imaging.

How much did OpenAI pay to acquire Glass Imaging?

OpenAI agreed to buy Glass Imaging for more than $300 million, according to a Wall Street Journal report. The startup had previously raised roughly $30 million from investors before this acquisition deal.

What specific technical problem does Glass Imaging solve for smartphone cameras?

Glass Imaging uses AI to overcome the physical size constraints of smartphone cameras, specifically focusing on how to achieve good depth and low-light performance from a lens smaller than a coin. This involves solving complex problems in optics and computational imaging rather than traditional software development.

Why is OpenAI's acquisition of a camera hardware company significant?

OpenAI's purchase of Glass Imaging represents a shift beyond language models and chatbot features into hardware development and physical world perception. The acquisition suggests OpenAI is building toward hardware systems that can directly see and interpret the world, integrating specialized expertise in computational imaging into its broader technology stack.

LIVE00:23Anthropic CEO Calls for Third-Party AI Evaluators