Global demand for nutrition continues to increase, creating new economic pressures—and opportunities—for farmers. At the same time, the agriculture industry must contend with the push toward more sustainable practices.
The emergence of rapidly evolving technologies, such as AI, offers agriculture players another powerful tool to meet these challenges head on and unlock greater efficiency and effectiveness throughout their businesses. Generative AI (gen AI) in particular has captured the imaginations of many leaders in agriculture and beyond and could be the impetus to create significant change.1 It has also brought to light the application of many other, long-existing approaches, such as analytical AI, with proven use cases and still relatively low levels of adoption.
When combined, analytical AI and gen AI have the potential to unlock value across the value chain and across business operations. This article explains how companies in the $4 trillion global food production industry2 can comprehensively strengthen their AI efforts by leveraging gen AI. Doing so can create economic value3 in two key areas: first, on the acre by improving on-farm economics such as labor and input costs and yields, and second, for the enterprise through increased sales growth, productivity, and operational efficiencies. Overall, our analysis shows that AI can create $100 billion in the former area and $150 billion in the latter.4
Applying gen AI in agriculture
Generally speaking, “gen AI” refers to applications that process large and varied sets of unstructured data, including geospatial and weather data, and perform more than one task. In this way, gen AI can generate new ideas by identifying patterns in large unstructured data sets, particularly when it comes to complex tasks such as molecular research, marketing or agronomy, and code generation. By contrast, analytical AI typically solves specific tasks by making predictions based on well-structured data sets and predefined rules. Examples here include forecasting sales, segmenting customers, and conducting sentiment analysis.
Agriculture is particularly well suited for disruption by AI and gen AI because of its high volumes of unstructured data, significant reliance on labor, complex supply chain logistics, and long R&D cycles, as well as the sheer number of farmers who value customized offers and low-cost services. As an example, gen AI can develop testing scenarios by synthesizing millions of data points on weather, soil conditions, and pest and disease pressure, and analytical AI models can then simulate those scenarios. Using both technologies in tandem has the potential to increase efficiencies, lower costs, and improve environmental impact for all agricultural players.
The significant value at stake
AI can create significant value for agriculture in two key areas: 1) on the acre, which refers to crop and livestock production, and 2) for the enterprise, which refers to business functions.
On the acre
AI and gen AI can help optimize the use of inputs and manage labor efficiently. For example, gen AI–enabled virtual agronomy advisers, which mine data sets such as weather, soil conditions, and pest and disease pressure, can help farmers make better-informed decisions to improve yields (Exhibit 1).
Up to half of the value at stake will be driven by such solutions for better yield management. Additional value will be driven by reducing labor costs via autonomous solutions to enhance the existing workforce and reduce dependency on labor in operations, as well as by input cost savings via new insights and data handling for precision agriculture to optimize inputs and reduce waste.
It remains unclear which agriculture players will take the lead and create products and services that combine analytical AI and gen AI for farmers. Gen AI technologies are more accessible than ever before, which means there is increased potential for start-ups and new entrants to capture value on the acre from larger companies. In any case, players across the value chain will need to move quickly in the months and years ahead.
For the enterprise
Across use cases, the combination of analytical AI and gen AI can create additional value by driving functional efficiency gains. The majority of this value will likely be enabled by analytical AI and complemented by solutions that are enabled or enhanced by gen AI. On this point, many organizations have historically focused their AI solutions on support functions, while our research shows that analytical AI and gen AI use cases add the largest value to core functions, such as R&D and products, marketing and sales, agronomy and sustainability, and operations (Exhibit 2).
The following examples illustrate how the combined forms of AI can create additional opportunities for players at every step in the value chain, even as relative distribution across business functions may differ:
- Input players. Many seed and crop protection players rely heavily on innovation, and analytical AI and gen AI can be used to enhance the full R&D life cycle, from research and discovery to development and product launch. In research and discovery, gen AI can help generate initial hypotheses by conducting a natural language scan of patents and scientific research, or screen large sets of genomic data to propose target sequences for crop innovation. Foundation models trained on specific modalities such as genomic, proteomic, or small-molecule data can help prioritize hypotheses based on end-state properties such as drought or pest resistance for genetically modified crops or improved efficacy and sustainability of pesticides (Exhibit 3). These tools are then built in an active learning loop in which models recommend hypotheses for testing in the lab, and the resulting data accelerates self-improvement. And in product launch, gen AI can accelerate product registration by automating data collection and analysis, generating documents, and providing insights into the regulatory landscape (for example, by monitoring changing regulatory procedure requirements).
-
Input distribution and growing and production services. Players that serve farmers directly—whether through sales of agriculture inputs or through services such as agronomic advisory, finance, insurance, and precision agriculture—serve millions of farmers who value customized offers and low-cost services. Analytical AI and gen AI can support marketing and sales domains across pricing and margin management, customer service and experience, sales growth and productivity, and personalized marketing.
In pricing and margin management, analytical AI models can microsegment customers and generate price recommendations based on historical willingness to pay, while gen AI can monitor real-time demand, supply, and regulatory shifts to adjust price recommendations. In customer service and experience, gen AI can provide text responses as a first point of contact in the purchasing journey, helping customers with questions on product selection or sample ordering. It can also offer ongoing product support with automated troubleshooting, synthesizing across large data sets of potential issues and resolutions. In sales growth and productivity, gen AI can act as a sales copilot by generating personalized customer sales scripts, synthesizing across product value proposition and customer buying factors, while analytical AI can mine structured transaction data, helping sales reps prioritize the best next steps. Finally, gen AI can be used to create personalized and real-time marketing content based on the history of interactions unique to each customer, while analytical AI can develop and evaluate targeted offers based on value analysis (Exhibit 4). Such use cases can reduce the cost of creating marketing content and drive revenue growth by improving lead conversion with better customer identification.
-
Trade and primary processing. For players in this step of the value chain, analytical AI and gen AI will likely have the largest impact on operational excellence.
In procurement, gen AI can create initial drafts of request-for-proposal documents, synthesize contract terms to identify sources of value leakage, and generate new negotiation strategies. In supply chain, gen AI can help monitor and identify potential disruptions, such as fluctuations in the weather or changes in global trade flows, as well as enhance network optimization by generating new scenarios around SKU-level analysis and costs. In manufacturing, gen AI can act as a virtual subject matter expert by synthesizing operational manuals. It can also provide insights on specific problems encountered during production, enhance predictive maintenance and failure monitoring, and create optimized production schedules.
As previously mentioned, such use cases are merely illustrative examples. When it comes to implementation, use cases will need to be tailored to the specific needs of players in each step of the value chain.
Rewiring to capture value from AI
Although many agricultural players are excited about the possibilities of gen AI, the majority have not yet realized the full potential of their ongoing AI efforts. Harnessing this potential will likely require players to rewire their approach to the following AI and digital efforts5:
- Start with alignment on value with a business-led strategy. C-level officers and board members can take an enterprise-wide view to focus resources, use the right governance models to streamline investments, and impart agility and accountability to the organization. Doing so will likely entail transforming domains with clarity on source of value and differentiation. Today, many players are inundated with tech-forward approaches, bottom-up pilots, and ad hoc partnership models, each of which eventually drains resources and doesn’t deliver clear impact. A shift toward a top-down view can help transform domains (as opposed to use cases), which can create substantial value.
- Modernize tech infrastructure for AI. A careful technology strategy is critical to avoid fragmentation and ever-increasing infrastructure costs. Depending on the capabilities required to transform a domain, players can make the right choices in technology infrastructure, analytical models, build-versus-buy decisions, and establishing key third-party relationships with software vendors, cloud-service providers, and vendors of large language models. Many of the underlying components can be reused to enable scale-up versus buying monolithic applications.
- Double down on the data foundation. Gen AI has opened up new avenues to access large amounts of unstructured information, which has historically been difficult for agricultural companies to integrate. When it comes to deployment, leaders will likely need to make decisions about enterprise data architecture (data stores and embedding), data integration patterns (APIs), and infrastructure (the cloud) to manage both structured and unstructured data.
- Upskill talent within business and technology organizations. Implementation and value capture require the right talent and capabilities. Agriculture players can focus on building talent by upskilling internal workers and bringing in external talent. At-scale training plans should be linked to formal mechanisms, such as compensation and career progression. Historically, talent retention has presented a challenge because many roles have unclear trajectories. As AI is democratized, company leaders can build expertise and deepen roles with the right career paths and progression, including data stewards and engineers, software architects or developers, MLOps engineers,6 and legal or security experts.
- Plan ahead to manage risk. Risk has shifted left with gen AI, which means issues need to be identified and addressed earlier in the development life cycle. In turn, players will need to define, codify, and assess ethical, legal, and regulatory risks up front in the process by monitoring cybersecurity and consumer privacy. Gen AI has further exacerbated these challenges, and legal and risk teams have become critical parts of agile delivery, with clear frameworks and controls emebedded in the tech infrastructure.
- Embed an agile operating model. AI development is inherently iterative, which means organizations should have mechanisms in place to rally behind high-value domains, particularly when pruning nonvalue-added pilots. Traditional annual budgeting and funding models can limit an organization’s ability to shift resources based on learnings, and leaders can explore innovation-funding mechanisms to jump-start and subsequently absorb AI efforts in a steady state. In addition, an integrated squad with a clear business owner can help ensure that AI and product teams deliver on critical business problems.
- Prioritize adoption and change. All too often, players allocate the majority of their investment in AI to technology development, overlooking adoption and change management. This has been a critical challenge for agriculture, both in traditional agtech applications and in ongoing digital efforts. Given that gen AI can dramatically shift work processes, adoption and change should be critical components of strategy. However, these efforts will likely require players to reimagine new ways of working, upskill the workforce to behave differently, and carefully manage change and capability building.
Agriculture stands poised to achieve remarkable advancements in food production and operations alike, transforming the way crops are grown, harvested, and distributed while empowering enterprises to work more efficiently and make smarter decisions for a more sustainable future. Striking the balance between sustainable practices and economic pressures will become increasingly important in the years to come, and the combination of analytical AI and gen AI will likely play a key role in shaping the future of the industry.
