Bison Ventures Closes $160 Million Fund 2 for Physical-World AI
Bison Ventures closed a $160M Fund 2 to back AI, robotics and energy startups for the physical world, targeting what it calls a $50 trillion market opportunity.
By Amara Osei
3 min read
Updated

What's News
- Bison Ventures closed a $160 million Fund 2, following its $135 million debut fund.
- The firm targets 'Real World Intelligence' — a category it values at $50 trillion across the global economy.
- Portfolio company Cobot's robots have logged over 12,600 production hours and moved more than 40 million pounds of material.
- Great Sky, a Fund 2 recipient, claims chip efficiency improvements of up to one million times versus traditional GPUs.
- Thinking Machines secured a gigawatt-scale partnership with NVIDIA.
Bison Ventures has closed a $160 million Fund 2 to back founders building artificial intelligence and other breakthrough technologies for the physical world. The early-stage venture firm announced the close and said the fund will invest from pre-seed through Series B.
The new vehicle follows Bison's $135 million debut fund. Founded in 2021 by Tom Biegala and Ben Hemani, the firm concentrates on what it calls Real World Intelligence — applying AI to physical systems across manufacturing, healthcare, energy, robotics, infrastructure, and advanced computing. Bison puts the market opportunity at $50 trillion across the global economy.
What is the firm's thesis?
Bison's investment thesis holds that AI's next major wave of innovation will extend beyond digital applications into industries that produce goods, develop medicines, generate energy, and maintain essential infrastructure. The firm's team pairs venture capital experience with engineering expertise in robotics, semiconductor manufacturing, biology, materials science, energy, and infrastructure. That technical background, the firm says, helps it identify emerging technologies and support founders in complex scientific and industrial markets.
Founding Partner Ben Hemani laid out the timing argument in a statement: "We are still early in unlocking the full value of AI to help us build a better future in the physical world. There will be a rapid acceleration in the use of AI to transform how we manufacture affordable goods, develop effective drugs, and produce sustainable energy, all driving better outcomes for people and the planet. We saw this inflection point proved out in Fund 1's thesis, and we're already deploying Fund 2 to back the next generation of deeply technical founders creating entirely new categories in massive, well-defined markets."
Co-founder Tom Biegala framed the firm's selection criteria: "Five years ago, we started Bison to back the companies that may appear contrarian today but will look inevitable in hindsight. A new technology has to change value created by an order of magnitude, and when it does, it's incredibly hard to displace. Our success in Fund 1 has only reinforced our conviction in backing these step-change solutions."
What does the existing portfolio show?
Bison's current holdings illustrate the range of technologies underpinning the strategy:
- Zyphra is developing frontier AI models and infrastructure on AMD's full-stack computing platform.
- Cobot has deployed collaborative robots across healthcare, manufacturing, and logistics. Its robots have accumulated more than 12,600 production hours and moved more than 40 million pounds of material.
- Vivodyne launched what it describes as the world's largest human biological datacenter.
- Thinking Machines released its first open-weights model and struck a gigawatt-scale partnership with NVIDIA.
The numbers from Cobot are the firm's most tangible proof point: 12,600 production hours and 40 million pounds of material moved by machines working alongside people in hospitals, factories, and warehouses.
How is Fund 2 already deployed?
Bison reports progress at three companies already receiving Fund 2 capital. Grid Aero has developed a full-scale autonomous cargo aircraft. Great Sky is commercializing superconducting research through chips designed to deliver computing efficiency improvements of up to one million times compared with traditional GPUs. Inner Logic has demonstrated that surgical AI trained entirely through simulation can transfer to actual patient anatomy, following success in autonomous gallbladder removal.
The selection signals where the firm sees returns. A million-fold efficiency claim in chip design and autonomous surgery validated in live anatomy sit far from incremental software improvements — exactly the "step-change" Biegala describes as his filter.
Why does it matter?
Fund 2 reinforces Bison's focus on companies capable of delivering substantial technological improvements rather than incremental advances. The strategy emphasizes opportunities where breakthroughs in AI, computing, materials, and automation can reshape established industries and create new markets.
With $295 million now raised across two funds since 2021, Bison is positioning itself to capitalize on the growing convergence of AI and physical infrastructure, as advances in computing and autonomous systems open new industrial and scientific applications. The bet: the next wave of AI value gets built in factories, hospitals, and power grids, not just in software.
Source: GN: Venture Capital
More from Amara Osei
Show full bio
Senior reporter covering companies, deals and strategy in entrepreneurship, startups and venture funding at Business Bearings.
1,084 articles