As energy demand continues to rise with the growth of data centers and Physical AI, innovative solutions are being sought for which Velaura AI has entered the market to provide.

Velaura AI has raised $110 million in Series A funding to accelerate development of ultra-low-power computing technology designed for artificial intelligence data centers, robotics, drones and autonomous systems.

The Santa Clara, California-based AI infrastructure company said the financing brings its valuation to more than $1 billion. This gives Velaura unicorn status at a time when investors are focusing on the AI industry’s pressing challenge of meeting the growing demand for electrical power required to support expanding computing.

Seligman Ventures led the Series A round, with participation from new investor Capricorn Investment Group. Existing investors include Mayfield, Maverick Silicon, MARA, Premji Invest, Samsung Catalyst Fund and StepStone Group.

Velaura will use the capital to accelerate development and commercialization of its AI compute portfolio. Also, it will expand its engineering and customer-facing teams and deepen relationships with customers and strategic partners developing next-generation AI infrastructure.

AI’s Growing Power Problem Creates an Opportunity for Velaura

The rapid expansion of generative AI, large-scale reasoning models and other compute-intensive technologies has triggered massive investment in data centers. Hyperscale technology companies are spending hundreds of billions of dollars to expand AI data center capacity.

But adding more AI computing capacity increasingly depends on access to sufficient electrical power, cooling systems and supporting infrastructure. Electrical grid constraints and lengthy infrastructure development timelines are limiting how quickly new computing resources can come online.

Velaura is attempting to address that bottleneck at the semiconductor level.

The company’s technology is designed to increase the amount of AI computing that can be performed for each watt of electricity consumed.

Improvements in performance per watt could potentially allow data centers to increase AI capacity while reducing power consumption, cooling requirements, and overall operating costs.

“Every advance in AI, from reasoning models to embodied intelligence, creates demand for more compute, and ultimately more power,” said Rajiv Khemani, co-founder and CEO of Velaura AI.

Khemani said the next stage of AI development will depend not only on more capable models, but also on improved economics surrounding the computing infrastructure required to operate them.

Titan Core Targets Greater AI Performance Per Watt

At the center of Velaura’s technology strategy is its recently introduced Titan Core™ silicon platform, a proprietary digital chip intellectual property and design architecture.

According to the company, Titan Core can deliver a 2x to 4x improvement in performance per watt for mathematical operations used in AI accelerators while maintaining performance.

Velaura also points to the commercial history of the underlying technology as an important differentiator. The technology has previously been deployed in more than 30 million application-specific integrated circuits, or ASICs, manufactured using advanced semiconductor process nodes.

That serves as the company’s technology foundation.  It has already demonstrated manufacturing yield and reliability at commercial scale rather than remaining solely an experimental chip architecture.

Patrick Moorhead, founder, CEO, and chief analyst at Moor Insights & Strategy, said that improving performance per watt has implications that extend beyond electricity consumption.

Greater efficiency could help AI infrastructure operators reduce total cost of ownership, manage thermal constraints and deploy additional AI capacity within existing facilities.

Physical AI Expands the Market Beyond Data Centers

Velaura is also positioning its technology for what the industry increasingly calls Physical AI—artificial intelligence operating in machines that interact directly with the physical world.

That category includes intelligent robots, drones, autonomous machines, and other embodied AI systems.

Unlike large data centers, these systems frequently operate within strict limitations on battery capacity, heat generation, weight and physical space. That makes energy-efficient computing especially important.

Velaura’s ultra-low-power computing architecture can be extended into those applications. This potentially gives the company exposure to both the expanding AI data center market and the emerging robotics and autonomous systems industries.

“Physical AI represents one of the next major frontiers for AI, and it will require a fundamentally different approach to compute centered on extreme power efficiency,” said Umesh Padval, managing partner at Seligman Ventures.

Padval said Velaura represents Seligman’s first investment specifically focused on Physical AI.

Experienced Semiconductor Team Behind Velaura AI

Velaura’s leadership and engineering organization includes veterans of some of the technology industry’s most prominent semiconductor and computing companies, including Apple, NVIDIA, Google, Qualcomm and Marvell.

The team includes executives and engineers with experience in developing high-performance, low-power semiconductor platforms. This is in addition to scaling chip businesses capable of shipping products at high volume.

Mayfield Managing Partner Navin Chaddha said his firm has previously backed Khemani in three other ventures and has also worked with Velaura executive Manu Gulati.

The combination of established semiconductor expertise and growing demand for energy-efficient computing appears to be an important factor behind investor interest in the company.

Energy Efficiency Behind AI Infrastructure Growth

AI’s rapid expansion is transforming computing infrastructure. But the industry’s next phase may depend as much on energy efficiency as raw processing capability.

This is tied to data center developers facing rising electricity requirements. Also, robots, autonomous vehicles, and other intelligent machines perform increasingly sophisticated AI calculations within much smaller power envelopes.

That creates an opportunity for semiconductor companies to increase computing performance without a corresponding increase in energy consumption.

With its technology already validated across tens of millions of chips, Velaura AI is positioning itself at the intersection of two expanding technology markets: hyperscale AI infrastructure and Physical AI.

As Velaura translates its reported performance-per-watt improvements into widely adopted AI accelerator and autonomous-system platforms, energy efficiency could become an important competitive factor in determining how quickly AI infrastructure can scale.