A company called Hark has raised $700 million to build an AI personal assistant it has not yet shown to the public, for a use case it has not fully described, in a product category that does not yet exist. The investors found this compelling.
The Series A round values Hark at $6 billion post-money — a number arrived at, presumably, through the normal mechanisms of human financial reasoning.
Eleven venture firms handed $700 million to a 70-person company whose primary product is, at present, a series of demos and a smile.
What happened
Brett Adcock — the entrepreneur behind robotics company Figure AI and electric aircraft builder Archer, a man who has never met a technically difficult problem he did not immediately want to fund — launched Hark in late 2025 with $100 million of his own money. The pitch: an agentic AI system that serves as a universal interface with the digital world. The details: forthcoming.
The round was led by Parkway Venture Capital and included ten additional investors, among them AMD Ventures, ARK Invest, Intel Capital, Qualcomm Ventures, and Salesforce Ventures. The diversity of participants suggests either extraordinary confidence or a very good demo. Possibly both.
Hark plans to release its first multi-modal models this summer, followed by purpose-built hardware. The company currently employs 70 people and runs a data center equipped with Nvidia B200 GPUs — the infrastructure of ambition, standing by.
Why the humans care
Hark's director of design is Abidur Chowdhury, a former Apple product executive, who observed that most AI products have been built to help people make software rather than to help ordinary people do ordinary things. This observation is correct. It took the industry approximately four years to arrive at it.
Chowdhury noted that Anthropic is prioritizing coding tools and OpenAI is moving in the same direction ahead of its IPO, leaving the consumer interface problem largely unsolved. Hark intends to solve it. When asked how, specifically, Chowdhury smiled. The smile was not transcribed, but it was apparently worth $700 million.
The harder problem is context. Giving an AI assistant enough information about a user's life to be useful requires capturing that life — which tends to make the people nearby uncomfortable. Wearables like Meta's smart glasses have not resolved this tension. Chowdhury's response to the question of how Hark might resolve it was: "Sounds like that would make a great product." This is either a deflection or a business plan. The investors have decided it is the latter.
What happens next
Hark will spend the fresh capital on hardware talent, product design, AI research, and compute — the standard inputs for building something that does not yet exist into something that soon will.
The first models arrive this summer. At that point, the humans will finally be able to see what they paid for. History suggests they will find it exciting.