SAN FRANCISCO, Asia could play a bigger role in the growth of Physical AI, as investors look for ways to bring intelligent machines from prototypes into real-world use.
The region has established strengths in semiconductors, electronics manufacturing and industrial supply chains. These capabilities could help robotics companies develop and deploy AI-powered machines at scale.
The opportunities were discussed at the Silicon to Scale: A Physical AI Summit during SF Tech Week. Investors highlighted Taiwan, Japan and Southeast Asia as important markets for the emerging industry.
The event was co-hosted by Startup Island TAIWAN Silicon Valley Hub and SparkLabs Taiwan. It attracted more than 170 investors, industry leaders and startup founders. Five Taiwanese startups also showcased their work: Morale AI, MetAI, Jmem Tek, Neuroshine and APMIC.
Asia’s Manufacturing Strength Could Support Physical AI
Physical AI refers to artificial intelligence that interacts with the physical world. Examples include industrial robots, autonomous machines and intelligent systems used in factories.
Jeff Tao, Citi’s director of AI infrastructure, said Asia’s semiconductor and electronics supply chains could support the industry’s growth. He also highlighted Southeast Asia, including Malaysia, as an important base for semiconductor assembly and testing.
These existing capabilities could help companies diversify production as geopolitical uncertainty affects global supply chains.
Taiwan also has a strong position in semiconductor manufacturing and advanced electronics. These industries provide important foundations for the development of AI-powered robotics.
Meanwhile, Japan offers a different advantage. Kosuke Yusa, general partner and head of investment at Open Network Lab Fund, DG Incubation, said Japanese factories often need robots that can adapt to changing production requirements.
This flexibility is important because industrial robots must handle different tasks and operating conditions.
Access to Real-World Data Remains a Challenge
Despite Asia’s manufacturing strengths, investors identified access to real-world data as a major challenge for Physical AI.
AI systems that work in factories need more than digital information. They also require specialised data collected from production lines, machinery and other physical environments.
However, manufacturers may be reluctant to share this information because it can contain confidential operational details.
Vandon Duong, a principal at DCVC, said companies developing AI systems are willing to pay for diverse, carefully labelled datasets. This creates opportunities for startups that specialise in collecting and preparing data.
However, businesses in this market must also find ways to maintain a competitive advantage as more companies enter the field.
Computing Costs Create New Investment Opportunities
The summit also explored the cost of building and operating AI infrastructure.
Jeff Tao said electricity availability has become a growing constraint for AI infrastructure development in the United States. Earlier challenges included access to graphics processing units (GPUs) and data centres.
As a result, investors are examining technologies that can make computing more efficient.
These include improved power systems, advanced cooling technologies and more efficient semiconductor manufacturing processes.
Duong also highlighted opportunities in technologies that reduce inference costs. Inference is the process through which an AI model uses what it has learned to generate an answer or make a decision.
Lower computing costs could help startups deploy AI systems more efficiently.
Taiwan and Silicon Valley Could Build Stronger Links
The discussions highlighted opportunities for Asian startups to combine regional manufacturing expertise with American investment, customers and technology markets.
Edgar Chiu, founding managing partner of SparkLabs Taiwan, said investors in Silicon Valley are increasingly interested in working with Asian strategic investors as Physical AI moves towards commercial production.
He said SparkLabs Taiwan aims to connect Taiwanese founders with US customers and investors. It also seeks to link American companies with Taiwan’s supply chains and strategic investment partners.
Uly Su, director and head of Startup Island TAIWAN Silicon Valley Hub, said the organisation is arranging industry discussions and investor gatherings to help Taiwanese startups build international partnerships.
These efforts aim to strengthen the region’s position in the emerging Physical AI industry.
AI Growth Brings Opportunities and Concerns
Although investors expressed confidence in the industry’s long-term potential, they also acknowledged concerns about high AI valuations and infrastructure costs.
The effects of automation on employment remain another important issue.
Tao warned that AI-driven productivity gains could displace some workers in the short term. This raises questions about how businesses and society will share the economic benefits of automation.
As Physical AI advances, Asia’s manufacturing expertise, access to data and international partnerships could help shape the industry’s next stage of growth.
Source: PR Newswire | Rewritten by: Minutesmy
