EIC’s new Advanced Innovation Challenges pilot draws 709 proposals across Physical AI and NAMs
- ›First EIC Advanced Innovation Challenges pilot attracted 709 proposals requesting €130.7 million.
- ›Physical AI drew 425 proposals while disruptive NAMs for safety assessment drew 284.
- ›Companies dominate applications at 69 percent with submissions from 39 countries.
- ›Selected teams are set to receive €300,000 for up to nine months, with a path to apply for up to €2.5 million in 2027.
- ›The pilot tests stage gated, demand driven deep tech funding aimed at faster market uptake.
A strong response to a new EIC pilot, and a test of demand driven deep tech funding
The first call under the European Innovation Council’s Advanced Innovation Challenges pilot drew 709 proposals across two themed topics and a total of €130.7 million in requested EU support. Submissions came from 39 countries with the largest volumes from Germany with 84, Italy with 71 and the Netherlands with 62. The majority of applicants were companies at 490 or 69 percent. Higher education institutions accounted for 124 or 17 percent and research organisations for 89 or 13 percent. The figures point to high interest from industry in a short runway instrument designed to bridge strong research to near term market testing.
Where the interest landed
| Metric | Number | Notes |
| Total proposals | 709 | Across two challenges |
| Total funding requested | €130.7 million | For feasibility and benchmarking activities |
| Countries represented | 39 | EU and associated participation |
| By challenge | ||
| Accelerating Physical AI | 425 | Embodied and edge AI systems that act in the physical world |
| Translating disruptive NAMs into practice | 284 | New Approach Methodologies for safety assessment |
| By applicant type | ||
| Companies | 490 (69%) | Largest share of applications |
| Higher education | 124 (17%) | Universities and similar institutions |
| Research organisations | 89 (13%) | Public and private research actors |
The distribution shows a clear skew toward Physical AI, consistent with growing investor and corporate demand in robotics, sensors, and cyber physical systems. The applicant type breakdown underscores the EIC’s pull on SMEs and startups. The reported counts and percentages do not fully sum to 100 percent, suggesting a small residual category such as public bodies or non profit organisations not detailed in the notice.
What the Advanced Innovation Challenges pilot is trying to prove
The pilot tests whether this format can move promising deep tech faster toward adoption than conventional research grants. It sits between exploratory research and full scale market readiness and is meant to complement rather than replace other EIC instruments.
Inside the two challenge areas
Next steps and indicative timeline
| Milestone | Timing | What to expect |
| Evaluation by independent experts | Spring 2026 | Remote assessment of proposals against call criteria |
| Applicants informed of results | By May 2026 | Indicative notification window |
| Lump sum project period | Up to 9 months | €300,000 per selected team to prepare and benchmark solutions |
| Follow on opportunity | 2027 | Most promising projects may apply for up to €2.5 million grants |
The 2027 follow on grants are positioned as an opportunity rather than a guaranteed step, which is consistent with a stage gated approach and should be read as a second competitive filter. Details on the available budget for either stage were not disclosed in the notice.
Signals and caveats from the numbers
The headline figure of 709 proposals confirms pent up demand for short cycle, adoption oriented deep tech support. The total requested €130.7 million across all submissions indicates applicants are calibrating ask sizes to the nine month scope. Without the published call budget or target number of awards, it is not possible to infer likely success rates. Given typical EIC oversubscription patterns, significant attrition is likely.
The dominance of company led applications suggests readiness to engage with market tests, but projects in Physical AI and NAMs face non trivial hurdles that often exceed nine month horizons. Certification, validation, and integration with existing infrastructures can extend timelines. There is also execution risk if demand side partners are named in proposals but not tightly committed to co development activities.
How this fits into the wider EU innovation toolkit
Practical takeaways for applicants and observers
| Parameter | Value | Implication |
| Grant size | €300,000 lump sum | Favors tight workplans and measurable benchmarking |
| Project duration | Up to 9 months | Focus on validation and demand engagement rather than full product build |
| Follow on path | Up to €2.5 million in 2027 | Second stage competitive application, not automatic |
| Evaluation | Independent experts | Strong emphasis on feasibility, adoption potential and team capability |
| Demand engagement | Encouraged early | Letters of intent and sandbox access strengthen credibility |
For Physical AI, credible testbeds, safety cases and performance against standard benchmarks matter. For NAMs, alignment with regulatory validation frameworks and evidence paths is decisive. Across both, early customer or regulator involvement can de risk adoption claims but must translate into time bound commitments, not general endorsements.
What happens next
Evaluation now shifts to the expert panels. Notifications are expected by May 2026. Selected teams will enter short sprints with €300,000 to demonstrate feasibility and benchmark against use cases. Only a subset will be invited to compete for up to €2.5 million follow on grants in 2027. Until budgets and selection volumes are published, stakeholders should treat the pilot as a high competition gateway to sharpen propositions and forge demand side traction rather than as a guaranteed bridge to scale funding.