EAFRD and the CAP Strategic Plans — how EU farm funding reaches digital and AI adoption, and why no euro figure for it exists
Europe (all EU member states; 28 approved plans, Belgium having two)
Content
The July 2026 funder scan named EAFRD and left the per-member-state picture as a gap (G-206). This unit closes it, and the finding is uncomfortable for anyone who wants a single number: the EU’s largest farm-level funding instrument has no published euro figure for digital or AI adoption at all. What it has is a 274,000-farm uptake target, seven member states with precision-farming eco-schemes, and a Commission report saying the CAP is “not always the primary driver of digitalisation, but one contributor”.
Headline. EAFRD is €95.5bn for 2021-2027 (including €8.1bn of NextGenerationEU money), flowing through 28 national CAP Strategic Plans carrying €264bn of EU funds; digitalisation is a cross-cutting objective with no budget line of its own; the only quantified EU target is that “more than 274,000 farms are to be supported by digital farming technology”; and the corpus’s contested claim that EAFRD is the largest single EU funder of digital agriculture deployment is positively counter-evidenced.
The money, verified 2026-09
The Commission states: “€387 billion in funding will be allocated to the CAP for the 2021-27 period… the EAGF, which has been set at €291.1 billion (in current prices); and the EAFRD, which will amount to €95.5 billion.” The financing page gives the total as €386.6bn. The €95.5bn is not a clean rural-development envelope: it “includes an injection of €8.1 billion from the next generation EU recovery instrument” (roughly 8.5%), and the in-plan EAFRD EU contribution across the 28 approved plans is €66bn, rising to €109bn of total public expenditure once the average 40% national co-financing is counted.
The 28 approved plans (Belgium has two) carry €264bn of EU funds for 2023-2027 and €307.35bn of total public expenditure. EAFRD by intervention type, EU contribution: agri-environment-climate commitments €20.29bn; areas with natural constraints €10.60bn; investments €18.43bn; setting-up/start-ups €3.41bn; risk management €2.73bn; cooperation €7.03bn; knowledge and information €1.13bn; technical assistance €1.86bn.
How digital is coded. Regulation (EU) 2021/2115 makes digitalisation a cross-cutting objective and part of Specific Objective 2 (competitiveness), and obliges member states to draft digital strategies (Art. 114(b)). Farm advisory services must cover digital technologies (Art. 15(4)(f)) and nutrient management via digital applications (Art. 15(4)(g)). Wine-sector interventions explicitly include “tangible and intangible investments in innovative processes for the digitalisation of the processes” (Art. 58(1)(e)). The mechanisms a member state can use: investments (Art. 73/74) for broadband, precision-farming equipment and digital technology in holdings; eco-schemes (Art. 31) and AECC (Art. 70) rewarding precision input optimisation; sectoral interventions (Art. 49-60); farm advisory services including the Farm Sustainability Tool for Nutrients (FaST); cooperation (Art. 77) including EIP operational groups, LEADER and Smart Villages; and knowledge exchange (Art. 78) for digital-skills training and demonstration farms. There is no AI-named intervention and no AI budget line — AI is subsumed under digitalisation, robotics and precision agriculture.
The number that does not exist
No published EU-level euro figure exists for CAP spending on digital agriculture, precision farming or AI. The only quantified headline is uptake: “Overall, more than 274,000 farms are to be supported by digital farming technology.” The Commission’s own 2023 overview already observed that the national digital strategies “include a limited consideration of digital technologies as an enabling tool for other CAP objectives… and a limited focus on the development of digital skills.”
The authoritative recent EU document — the EU CAP Network thematic report Assessing digitalisation under the CAP Strategic Plans (November 2025) — declines to quantify a total and states:
“The CAP is therefore not always the primary driver of digitalisation, but one contributor within a broader framework of policies and investments.”
Where a member state has published a discrete figure, the verified example is Germany: “11 projects on optimising the data and information exchange between e.g. agricultural machines, vehicles or with central data processing have received financing of EUR 15 million over three years from early 2025” (OECD, Agricultural Policy Monitoring and Evaluation 2025, EU chapter). Greece supports producers “for the purchase of a digital application licence which include information on farms, soil sampling, management practices, crop history and yield targets”. Seven of the 28 plans (Belgium-Flanders, Czechia, Denmark, Greece, Ireland, Latvia, Sweden) plan a dedicated eco-scheme to compensate farmers for the use of precision-farming technologies.
Adoption evidence (not budget): the JRC study The state of digitalisation in EU agriculture: Insights from farm surveys (JRC141259, 19 September 2025) found 93% of surveyed farmers used at least one IT/software tool, 79% at least one crop-specific technology, 83% at least one livestock-specific tool, and 76% expected economic benefits. That is technology use of some kind, not AI deployment — the corpus should not read it as AI adoption.
Member-state detail
| Member state | CAP Strategic Plan facts | Digital/AI-relevant content |
|---|---|---|
| France | EAFRD envelope €7.297bn (2023-2027); approved 31 August 2022; six amendments, latest 2 March 2026 | PSN treats digital as a continuing priority and references robotics; national CAP budget ~€9bn/year |
| Spain | PEPAC €47.724bn of EU funds for 2021-2027, >€50bn with co-financing; approved 31 August 2022; four amendments, latest 12 December 2025 | cooperative-federation emphasis (see the Spanish cooperative cluster units) |
| Netherlands | Approved 13 December 2022; five amendments, latest 25 November 2025 | NSP states the country “will focus on the development of smart agriculture, among other things through further development of precision farming, digitalisation”; Dutch ministry co-supported the EU CAP Network’s robotics/AI seminar |
| Germany | GAP-Strategieplan ~€30bn EU funding; approved 21 November 2022; four amendments, latest 5 December 2025 | the €15m / 11-project interoperability programme from early 2025 |
| Italy | PSN PAC Chapter 8 “Modernizzazione AKIS e Digitalizzazione”; digitalisation strategy of 3 aims, 10 strategic lines, 22 actions; approved 2 December 2022; five amendments, latest 27 November 2025 | census baseline: fewer than 16% of Italian farm businesses use digital tools; ~33%/16% Centre-North; 7% in the South |
| Poland | Plan Strategiczny WPR 2023-2027, ~€25.2bn (secondary source); KSOW+ network launched 1 July 2023 | advisory centre CDR publishes digital/AKIS analyses |
| Ireland | one of the seven precision-farming eco-scheme states; CSP total >€9.8bn (secondary); Knowledge Transfer Programme; EIP funding €17.8m to 11 projects | Knowledge Transfer budget not verified against a primary DAFM document |
Correction to the July scan: naming France, Spain and the Netherlands as the “AI-relevant” member states is not wrong but incomplete — Germany, Italy, Greece, Ireland and the seven precision-eco-scheme states all carry verifiable digital content.
Advisory and knowledge systems (AKIS)
The EU CAP Network (which replaced the ENRD, EIP-AGRI and the Evaluation Helpdesk) runs a Digitalisation section covering “digital and data technologies such as smartphones, tablets, in-field sensors, drones, and satellites. Robots and AI technologies are not left behind”. Named AI-relevant activities:
- the EU CAP Network seminar “Robotics and Artificial Intelligence in farming and forestry”, Utrecht, 19-20 February 2025 (event report 4 July 2025; factsheet 12 June 2025);
- AKIS evaluation guidelines (May 2023) counting “digital innovations supporting knowledge flows” and “training programmes completed on digital farming technologies”;
- the November 2025 thematic report’s three-tier classification system, ending in activities such as “sensors and monitoring equipment, GPS-guided machinery, variable-rate application systems, farm management software, robotics, … data-sharing platforms”.
What changed since July 2026
- 8 September 2026 — agri-food innovation strategy. Communication COM(2026)453 sets out a new Strategic Approach to Research and Innovation uniting AgRI 2040 (agriculture, forestry, rural areas) and Food 2040, aiming to “accelerate the uptake of knowledge and innovation” and to help “startups, scaleups and SMEs… move promising innovations closer to the market”.
- CAP post-2027 / MFF 2028-2034 (proposed 17 July 2025): a single fund replacing the two pillars, integrated into National and Regional Partnership Plans, an NRP envelope of €865bn, ring-fenced ≥€300bn for income and crisis support, €453bn for national needs, and €40bn under a “Health, Biotech, Agriculture and Bioeconomy” window (€19.7bn Horizon Europe + €20.4bn Competitiveness Fund); “successful tools like LEADER will continue”.
- CAP simplification: proposal 14 May 2025; Regulation (EU) 2025/2649 adopted December 2025 to restrict Commission approval to strategic amendments.
- CAP 2028-2034 guidance (28 April 2026) notes that in budget year 2025 “around 28% of total public funding under CSP was allocated to eco-schemes and agri-environmental and climate commitments”.
- Also post-dating the July snapshot: the JRC farm-survey study (19 September 2025) and the EU CAP Network digitalisation report (November 2025).
What this unit is doing in the taxonomy
Anchors the EU farm-level funding layer — the money that reaches holdings rather than consortia or startups. Distinguishes from:
- Horizon Europe Cluster 6 (
units/horizon-europe-cluster-6-food-bioeconomy-agriculture.md) — research consortia; EAFRD flows to farms and advisory systems. - EIT Food (
units/eit-food-eu-knowledge-innovation-community.md) — startup accelerator; EAFRD does not fund startups directly. - Copa-Cogeca (
units/copa-cogeca-eu-farmer-cooperative-federation.md) — the constituency whose members receive this money. - EU Mission Soil (
units/eu-mission-soil-deal-for-europe-living-labs.md) — project-based Living Labs; CAP is the mainstream farm-level channel.
Why it matters for talks
- The absence of a number is the finding. The EU spends €95.5bn on rural development and cannot state what share goes to digital, let alone AI. Any talk asserting “the EU spends X on agritech AI through the CAP” should say this instead.
- The Commission’s own framing is the strongest available quote for the adoption argument: the CAP is “not always the primary driver of digitalisation, but one contributor”.
- 274,000 farms is the EU’s only quantified digital-adoption target — a usable, official figure for talks that need scale without pretending it is money.
- Italy’s 7% South versus 33% Centre-North is the corpus’s clearest EU-internal digital divide and belongs in any talk about digital equity within Europe, which otherwise tends to be framed North-South at the global level.
- Seven member states now pay farmers to use precision farming. That is the EU’s most concrete demand-side mechanism for agritech adoption, and it is a policy fact rather than a vendor claim.
- The post-2027 architecture is a single fund with partnership plans — worth naming in any talk about where EU agrifood AI money goes next.
Critical context
- EAFRD’s headline €95.5bn includes €8.1bn of NextGenerationEU recovery money; the in-plan EU contribution is €66bn.
- No EU-wide euro figure for CAP digital spending exists; treat any such number as unsourced (G-418).
- The JRC 93%/79%/83% figures are technology-use survey results, not AI deployment evidence.
- Member-state figures for France, Poland and Ireland rest partly on secondary sources; CSP factsheets and national plan PDFs should be read directly to confirm exact digital-intervention lines.
- 99.3% of the CAP budget is under shared management, which is why no single EU-level digital figure can be produced.
- The 274,000-farm target is a plan-based uptake target, not an achieved count.