Bedrock Robotics raises $600 million to put autonomous excavators on commercial job sites
San Francisco-based Bedrock Robotics has closed a $600 million Series C at a $4.1 billion post-money valuation, co-led by CapitalG and Emergence Capital with NVIDIA, Temasek and Bessemer among the backers. The company retrofits heavy construction equipment with autonomy kits and has now raised nearly $1 billion.
Bedrock Robotics, a company founded by former Waymo engineers to automate heavy construction equipment, has raised $600 million in Series C funding at a post-money valuation of $4.1 billion. Alphabet's growth fund CapitalG and Emergence Capital co-led the round, with new investors including the US Innovative Technology Fund, Bessemer Venture Partners and Temasek, and existing backers such as 8VC, Valor Equity Partners, Eclipse and NVIDIA participating.
The company's approach is retrofit rather than redesign. Instead of building new machines, Bedrock fits sensors, compute and control systems to existing excavators and other site vehicles so that contractors can run them autonomously on routine earthmoving tasks. The new capital is earmarked for expanding the number of Bedrock-equipped machines on commercial construction sites across the United States.
The raise brings total funding to nearly $1 billion, only eight months after a $270 million Series B in February. It was one of two $600 million-plus rounds in the same week, alongside Korean chipmaker FuriosaAI's $620 million Series D, in what trackers recorded as a $2.4 billion week for AI funding globally.
Construction and mining have become the most active frontier for physical AI because the environments are controlled, the labour shortages are acute and the economics of a machine running extra shifts are straightforward to prove.
Why it matters
Autonomy is moving from passenger vehicles, where the regulatory bar is highest, into industrial equipment where deployments can scale with less friction. Investors are now willing to fund physical AI at valuations once reserved for software, a sign that the technology is producing measurable site-level returns.
Indicops view
India's infrastructure pipeline, from highways to data centres, faces exactly the labour and productivity pressures that retrofit autonomy addresses, and the cost of Indian construction equipment makes the retrofit model particularly attractive. We expect public-sector infrastructure agencies to begin specifying autonomy-readiness in equipment tenders within a few years, and we advise clients to start collecting site telemetry now so that future automation has data to work with.
More in Robotics
Back to the newsroomDeepMind launches Gemini Robotics 2.5 with 96 per cent success in Mercedes-Benz parts packing
Google DeepMind has released Gemini Robotics 2.5, its most capable vision-language-action model, tested on over 200 tasks and rolling out via Agile Robots, Apptronik and Boston Dynamics. It reached 96 per cent success packing automotive parts in Mercedes-Benz trials. AlphaGenome 2 launched the same day.
Agility Robotics lists on Nasdaq as AGLT, the first US-listed pure-play humanoid maker
Agility Robotics began trading on Nasdaq on 5 November under the ticker AGLT after merging with Churchill Capital Corp XI. The deal valued the humanoid maker at $2.5 billion pre-money and raised about $465 million in gross proceeds, including a Foxconn-led private placement.
Humanoid robotics programmes move from demo to deployment pilots
Several manufacturers have begun limited factory pilots, with foundation models for control proving the enabling technology.