The China Mail - Europe should focus on industrial AI, SAP says

Artificial intelligence for use in industry is "where I believe Europe can succeed big time around the globe", Thomas Saueressig, a board member at the continent's biggest software maker, told AFP in an interview.

This was because of "the industrial knowledge and the industrial data and competency we have in Europe", he said at the company's headquarters in Walldorf, southwest Germany.

Europe is seen as lagging behind the world's top two economies in the strategic sector: the US with major players from OpenAI to Google, and China, which has a buzzing AI scene and hosts influential labs such as DeepSeek.

But fraying ties with Beijing, which Europe accuses of engaging in unfair trade practices, and the United States under President Donald Trump have given the continent fresh impetus to step up its efforts.

Industrial AI applications are typically smaller and less energy-intensive than the large models produced by major players, and designed for specific tasks that could range from developing construction plans to optimising electrical wiring in devices.

That is where "we can differentiate ourselves in Europe" by "leveraging the industrial expertise, the data expertise for specific models", said Saueressig, SAP's chief customer officer.

"We don't need to think about only the large language models, we also need to talk about the specific industry models."

- 'New potential' -

High-tech processes are nothing new on factory floors but AI promises to turbocharge them, and there are already signs efforts are being ramped up in Germany, Europe's biggest economy.

Last week, auto giant BMW unveiled a pilot project to deploy two AI-powered humanoid robots in a German factory this year to help in the production process.

"Advances in AI system performance unlock new potentials for automation," said Milan Nedeljkovic, who is BMW's head of production and is set to take over as CEO in May.

The carmaker has been building up its AI capabilities for some time and was moving "towards autonomous AI which can be used in manufacturing to take its own decisions", Nedeljkovic told journalists, including from AFP, in Munich.

German telecoms giant Deutsche Telekom and US chip juggernaut Nvidia recently launched an industrial AI hub aimed at helping European companies use the technology in processes ranging from design to robotics.

The hub says it offers a "sovereign AI platform" that can reduce "risky dependencies" on Chinese and US tech.

Europe's drive to boost its AI capabilities ties in with broader efforts to ensure "digital sovereignty", by dialling back a heavy reliance in particular on US tech giants and ensuring Europeans' data is stored safely at home.

Saueressig said SAP, which makes a wide range of business software, had seen "exponentially increased demand" for products that guarantee "digital sovereignty".

But the executive, who oversees the group's cloud computing operations, said some concerns about reliance on US tech were "overdone" and urged Europe to focus on producing world class technology so it could compete globally.

- AI disruption -

SAP has faced worries that new AI products could threaten its business, with its share price falling heavily in recent months.

But Saueressig says the group, which has been developing AI tools for its clients, can benefit as the technology provides "growth opportunities and more value for our customers".

Europe faces many challenges when it comes to building up its industrial AI offerings, experts warn.

These range from lower data centre computing power compared with the US and China, challenges in the field from manufacturing powerhouse China, and a lack of funding for promising start-ups.

Despite the challenges, the continent has plenty of opportunity and the AI race is in its early stages, said Antonio Krueger, head of the German Research Centre for Artificial Intelligence (DFKI).

"We don't know exactly where this will end and when it will end," he told AFP.

"It is not justifiable to claim that the race has been lost."

Z.Huang--ThChM