OpenAI just spent $300 million on a camera company most people have never heard of. According to TechCrunch, the ChatGPT maker has acquired Glass Imaging, a Los Altos-based startup that uses AI to dramatically improve smartphone photo quality. For a company that started as a research lab, OpenAI is starting to look a lot like a consumer hardware company.
Glass Imaging was founded in 2019 by Ziv Attar and Tom Bishop, two former Apple engineers who led the team behind Apple’s Portrait Mode. That’s not a minor credential. Portrait Mode changed how millions of people think about phone cameras, and the duo applied that same thinking to their startup. Glass Imaging’s approach is technically distinct from typical AI photo editing: rather than cleaning up an image after it’s been captured, the company trains neural networks on specific camera hardware to produce better photos at the moment of capture. It understands the physical quirks of individual camera modules across different phone models and compensates in real time.
The startup had raised around $30 million before this acquisition, making a $300 million exit a significant return for its backers. But the more interesting number is what it tells us about OpenAI’s priorities right now.
This is the second major hardware-related acquisition OpenAI has made in recent memory. In 2025, the company bought Jony Ive’s device startup io for $6.5 billion, a deal that confirmed OpenAI and the legendary Apple designer were building a physical AI product together. Rumors have since circulated that OpenAI is working on smartphones, earbuds, and companion devices. Glass Imaging fits cleanly into that picture. If you’re building a phone or a wearable with a camera, you want the best possible image quality baked in from the start, not bolted on later.
The competitive context matters here. Apple has spent years refining computational photography with its own silicon and software stack. Google’s Pixel phones have built a strong reputation on AI-driven camera processing. Samsung works closely with third-party AI imaging partners. OpenAI, if it ships a device, will be going up against all of them. Acquiring a team with deep expertise in hardware-specific neural imaging is a rational way to close that gap fast.
OpenAI did not comment on the deal. But the pattern of acquisitions tells its own story. This is a company building toward physical products, and it’s buying the technical talent to do it seriously.
