In this episode, we explore the simple idea that the usefulness of Artificial Intelligence (AI) in agriculture depends less on the technology itself, and more on whether people trust its outputs.
Across agriculture, use of AI continues to build, with tools for analysing weather, soils, crops, and markets becoming available, all of which supports faster and more informed decisions. Global conversations, including those at the recent AI4Agri 2026 event, reinforce both the opportunity and the challenge. As Vincent Martin, FAO Office of Innovation’s Director noted, ‘The value of AI in agriculture will not be measured by how sophisticated our models are. It will be measured by whether farmers trust them’. This shifts the focus from technical capability to relationships and trust.
Trust, in this context includes confidence in the quality of the data, clarity around how outputs are generated, and alignment with local conditions and experience. It also reflects whether someone feels confident enough to act on the information within their own system. When any of this is missing, even accurate outputs won’t be used.
For enablers of change, this means our role isn’t just about sharing information but can also include brokering trust between AI systems and the people using them. AI can generate options, but it cannot fully account for the nuance of individual farms, local variability, or personal risk preferences. Our role is to help interpret, validate, and adapt options to fit specific contexts, through judgement, transparency, and relationship-building.
We can see what happens when trust hasn’t been established. For example, early precision agriculture platforms offering variable rate recommendations often failed to get farmers signed up because they couldn’t easily understand how these were generated. Similarly, some crop disease identification apps delivered inconsistent results under field conditions, reducing confidence and leading farmers to rely on their own observations instead. In livestock systems, decision-support tools for grazing or feeding have sometimes been perceived as too generic, meaning they aren’t really helpful or relevant.
These examples point to a common pattern where lack of transparency, limited contextual fit, or inconsistent performance undermines trust. This aligns with another key insight from Vincent Martin, who stated that ‘Artificial intelligence is not simply accelerating agricultural research. It is exposing the inefficiencies in how we generate, share, and apply knowledge’ (FAO, 2026). When outputs are disconnected from how knowledge is actually used on-farm, trust breaks down and adoption stalls.
In contrast, trust grows when AI outputs are contextualised such as when a seasonal forecast becomes more useful because an advisor explains the data behind it, discusses the uncertainty of those forecasts, and relates it to local conditions. A nutrient recommendation gains credibility when it is adjusted to reflect paddock history. Pest and disease alerts become more usable when farmers validate them in the field and discuss them with their peers, building confidence through shared experience.
There is also a need to rethink capability development. Working effectively with AI requires a range of capabilities including technical expertise, systems thinking, facilitation, and digital literacy. It is not enough to just understand the tool. We should also be able to explain it, understand the inputs, contextualise the outputs, and support others to use it confidently. Or perhaps we form networks with people who have those capabilities so that together we can support farm advisors and farmers to use AI well.
Ultimately, AI does not replace the human dimensions of agricultural extension. We think it makes people even more important! So it comes down to whether people understand AI tools and outputs, see their relevance, and feel confident enough to act. When trust is present, AI becomes an enabler of better decisions. When it’s absent, even the most advanced and potentially useful tools will remain on the shelf.
We’ve shared our thoughts, now it’s your turn! Drop a comment below with your experiences, tips, and ideas on building trust and AI tools. We’d love this to be an ongoing conversation.
Thanks for reading this Enablers of Change episode! Don’t forget to subscribe to our newsletter to catch future episodes, and share this post with friends who might enjoy the discussion.
Until next time, all the best and keep building trust!
Resources
Food and Agriculture Organization. (2026). FAO wraps up its presence with contributions to AI4Agri 2026 in Mumbai. Available online.