Shaping the Deep Tech Revolution in Agriculture 2025

Conclusion Multistakeholder collaboration is essential to build ecosystems in which agri deep tech can thrive. Realizing the true potential of agri deep tech demands a candid acknowledgement of the significant challenges in scaling such technologies, particularly for smallholder farmers in emerging economies or agri small and medium-sized enterprises (SMEs). To effectively bridge the gaps in awareness, access and widespread adoption, a concerted, multistakeholder effort is imperative. Figure 17 outlines the key responsibilities of various stakeholders in fortifying these essential agri deep- tech ecosystems and enabling impact at scale. Stakeholder responsibilities in building agri deep-tech ecosystems FIGURE 17 Policy-makers — Establish policy frameworks: Develop foundational data gover nance frameworks, ensur e secur e access to data and incentivize data sharing to enable agri deep-tech innovations — Set up data-sharing infrastructure and data repositories: Establish data exchanges and r epositories that pr ovide foundational data for training AI models in agricultur e — Fast-track regulatory sandboxes and approval procedures: Implement regulatory sandboxes for novel technologies (e.g. CRISPR, nanotechnology) and cr eate transpar ent, predictable appr oval pathways — Encourage cross-country knowledge transfer: Facilitate dialogue and cr oss- country knowledge-exchange sessions Big tech firms — Champion open data: Facilitate widespr ead access to open r epositories for training, benchmark datasets and provide subsidized cloud computing infrastructur e with r obust cybersecurity protocols to pr omising innovators — Promote pre-competitive collaboration: Use expertise to foster cr oss-country technology transfer models and seed global technology partnerships, often on a pre-competitive basis — Build local capacities: Partner with domestic start-ups and academic bodies to contextualize solutions to local contexts befor e deployment — Identify fundamental agricultural challenges that accord with national priorities: Align ideation and R&D with national and subnational priorities to unlock gover nment support — Seek partnerships: Pr oactively seek interdisciplinary expertise and use regulatory/technical sandboxes for r eal- world validation — Re-engineer business models: Consider alter nate r evenue str eams stemming fr om business to business (B2B) models to over come challenges relating to limited ability and willingness to pay among farmers — Deploy demonstration farms/ experience centres: Establish places in which end users of technology can engage with a solution and lear n about its benefits to encourage adoption Deep-tech start-ups — Cultivate interdisciplinary talent: Integrate deep-tech modules into agricultural curricula and launch industrial exposur e initiatives — Lead foundational research on agri deep tech: Establish inter disciplinary deep-tech r esear ch hubs within agricultural universities and host academia–industry convenings — Disseminate knowledge and share data: Develop and contribute to open repositories of training and benchmark datasets, serving as vital public resour ces— Provide patient capital: Mobilize ventur e capital towar ds agri deep tech thr ough blended finance models and long-term patient capital vehicles — Underwrite technology risk: Build inter nal capabilities to understand and assess the unique technology risks of agri deep-tech innovations — Measure and report on impact: Consider establishing impact evaluation funds to assess and showcase the broader societal benefits alongside financial r eturnsFunders and investors Academia and r esear ch Source: Consultations with AI4AI community experts Shaping the Deep-Tech Revolution in Agriculture 37