Siemens Aims to Lead in "Industrial AI"

Siemens is setting its sights on becoming a frontrunner in the application of artificial intelligence (AI) for industrial development and manufacturing.

The Munich-based technology group aims to rapidly develop a large-scale foundational model for industrial applications--dubbed the "Industrial Foundation Model," Siemens Chief Technology Officer Peter Körte said Tuesday on the sidelines of a Siemens-hosted AI conference attended by around 700 participants. "Speed is the key. We want to be the first," Körte emphasized. "If we don't do it, others will."

The company is well positioned for this push, Körte added. "We have the resources, we have the network, but we also have the trust of our customers and credibility." Siemens has secured 2,000 AI-related patents and employs more than 1,500 experts in the field. Speaking to the Handelsblatt (Tuesday edition), Körte revealed that Siemens plans to invest a nine-figure sum in this effort.

On the sidelines of the conference, Körte voiced opposition to regulating AI in Europe. "Do we need an 'AI Act'? Absolutely not," he said. "We don't need more laws." Existing contracts with customers are entirely sufficient in the industrial context, Körte argued. "If customers notice something negative, they'll call us." With its strong industrial base, Europe is better positioned than the United States to take a leading role in the industrial use of AI, he said, adding that AI could deliver productivity gains of up to 20 percent.

However, Körte stressed that the key to success lies in the exchange of data between Siemens and its customers. Siemens will insist on corresponding agreements, as only then can the necessary accuracy for these applications be achieved, he said. Engineers need to feel that AI makes them faster, not slower--and that it is reliable. Otherwise, the software will quickly end up back on the shelf. The final authority will always remain with humans, Körte insisted. "We still believe there will be a human element."

Körte does not believe customers will hesitate to share their data in the future. Data from machines is less sensitive than medical data, where clinics and doctors are more cautious. For example, Apple may not reveal the design of its latest iPhone model, but it might share data that could help machines learn.

(Reporting by Alexander Hübner; editing by Ralf Banser. For inquiries, please contact our editorial team at [e-mail] (for politics and economic affairs) or [e-mail] (for business and markets).)