EU opens €80M funding calls to move European AI chips towards deployment
Written by Rebecca Uffindell Thu 1 Oct 2026
The Chips Joint Undertaking has opened two funding calls worth a combined €80 million (£68.2 million) to help European-designed AI chips move from working prototypes towards testing and deployment in European compute infrastructure.
The two calls cover different stages of that process. The first provides €10 million (£8.5 million) to turn advanced chip designs into working demonstrators that can be benchmarked. The second provides €70 million (£59.7 million) to establish the shared testing environment and procurement framework that could ultimately bring selected AI compute systems into European infrastructure.
Together, they create a pathway connecting European chip developers with the AI Factories, public high-performance computing (HPC) facilities and cloud providers that could eventually use their technology.
Both calls close on 27 October 2026 at 5 pm Brussels time, according to the Chips Joint Undertaking.
€10M to Turn AI Chip Designs Into Working Systems
The first call, AI chip demonstrators for EU compute infrastructure, has an indicative EU budget of €10 million.
Its purpose is to take sufficiently advanced European AI chip designs and turn them into working demonstrators.
In practical terms, that means producing more than the chip itself. The demonstrator combines the silicon with the circuit board and supporting infrastructure needed to operate and test it, including cooling, interconnects, and software.
Chips JU estimates an EU contribution of up to €1 million (£853,345) per project, although applicants can request a different amount.
Projects must include at least one EU-established AI chip design company responsible for the chip design and associated intellectual property. System integrators and equipment manufacturers can also participate where required.
The programme is aimed at relatively mature technology. Applicants need to demonstrate that their chip has already been taped out at a commercial foundry.
Put more simply, the design must already have reached the stage where it has been sent to a manufacturer for physical production. Engineering samples must either be available or expected within the required timeframe.
A printed circuit board capable of hosting the chip must also be in production or expected during the six-month development period.
Once completed, the demonstrators will undergo two months of performance testing on a common evaluation platform.
The strongest-performing technologies can receive an AI Compute Excellence Label. That label will be required to access a later programme intended to develop selected technology into complete rack-level systems.
That next stage, known as Topic A2, is not currently open and is not scheduled to launch in 2026.
€70M to Build the Testing and Buying Route
The second call addresses what happens around those technologies once they are ready to be evaluated.
The AI compute evaluation and deployment platform for EU infrastructure has an indicative budget of €70 million. Rather than funding another set of individual chip designs, it will establish a common way to test AI compute technologies and create a route for selected systems to be purchased for pilot deployment.
Applicants must form a buyers’ group involving organisations from at least three EU Member States or European Economic Area countries.
The group must include at least one organisation operating an EU AI Factory or comparable public AI/HPC facility and at least two EU-established cloud service providers.
The programme is expected to run for 42 months.
Initially, the infrastructure created through the call will benchmark the chip demonstrators produced under the €10 million programme.
Later, it is intended to test the complete rack systems developed through the future A2 programme.
The final phase moves from evaluation towards deployment. Selected systems would be purchased, installed and optimised in multi-rack clusters, rather than tested as isolated racks. This allows the technology to be assessed under distributed AI workloads closer to those it could encounter in operational infrastructure.
A Route From European Chip to European Infrastructure
The two calls therefore sit at opposite ends of a wider development programme.
The route envisaged by Chips JU can be understood as:
European chip design → working demonstrator → common benchmarking → rack development → rack-level testing → procurement → pilot deployment
Only the first demonstrator stage and the evaluation and procurement programme are currently open. The rack-development stage between them will follow later.
For European chip companies, that provides a potential route beyond demonstrating that a processor works. Successful technologies can be compared under common conditions and could eventually progress towards complete AI compute systems.
For AI Factories, public HPC facilities and cloud providers, the second call creates an opportunity to help shape how those systems are evaluated and subsequently purchased for pilot deployment.
The programme also has a wider European technology dimension. Its eligibility requirements deliberately involve EU-established chip designers, infrastructure operators and cloud providers, creating a route through which European-designed compute could reach European AI infrastructure.
The €80 million funding does not by itself establish European independence in AI compute. It does, however, address a practical part of that ambition: giving home-grown chip technology somewhere to prove itself and, if successful, a clearer route towards deployment at infrastructure scale.
Written by Rebecca Uffindell Thu 1 Oct 2026
Tags:
cloud infrastructure European Commission hpc semiconductorsMost Viewed News
Fri 2 Oct 2026
AI skills gap grows despite 64% of workers using AI