Agriculture’s Matrix: AI and the Future of Gender (In)equities

Farmers and AI 

Thanks to AI, farmers can analyze data including weather, temperature, water usage, and soil conditions in real time, which allows them to make better decisions and generate more bountiful yields. AI also helps farmers address labor shortages. Agricultural bots complement work done by human labor, and chatbots provide targeted advice to farmers seeking guidance with problems.  

But AI also has some potentially negative consequences longer term. Because AI is only available to those who can afford it, it could create winners and losers, with winners being those who can pay for, and benefit from, AI and losers being those who can’t. It could also dramatically change what the world looks like and how farms operate, forcing individuals who have historically relied on agricultural labor for income out of their jobs. More broadly, it is unknown how significantly AI might transform supply chains and what those shifts might mean for labor and daily life.  

In this post, I consider one of AI’s potential negative consequences: increased inequities between men and women farmers. I argue that if the current state of AI research and technology does not change, the existing inequities between men and women farmers will become starker. But there are also key ways existing work on AI could shift to not only prevent inequities from worsening, but also contribute to a more equitable agricultural landscape.   

Inequities between men and women farmers already exist. Worldwide, women tend to be employed for labor-intensive tasks, typically earn lower wages than men, and are more likely than men to be paid at piece rate. In half of the world’s countries, women are denied land and property rights despite laws being in place. For example, while a country’s laws may formally grant women rights to land ownership, customs and misogynistic social norms may undermine those rights, resulting in women not being allowed to hold land in practice. And where women do hold their own land, their plots are usually smaller, of an inferior quality, and have less secure rights than those that men hold.   

These inequities will only be exacerbated if AI in agriculture continues to follow its present trajectory—a trajectory that problematically excludes, and ultimately leaves behind, women farmers.   

Examining some of the leading companies in AI and robotics looking to transform the agricultural sector, it becomes clear that these transformations are excluding women’s insights. A rudimentary analysis of the Top 10 AI and Robotics Companies in agriculture indicates that men are primarily leading the way in the development of new AI technology for farmers:  

*Note: These counts are based on leaders’ names on the companies’ websites. No individuals specified their gender identities or pronoun preferences on the site, so the data were counted based on sociocultural norms regarding men’s and women’s names. To best capture individuals’ gender identities, and be inclusive of non-binary gender identities, companies would need to be contacted and surveyed. This, of course, would be a great avenue for future research. 

Of the companies that specify who sits in leadership roles, only one (Trace Genomics) includes women in leadership. None of the top ten companies include information about how the company is being responsive to existing inequities among farmer. And none of the companies feature women farmers using their technology. It is also concerning that four of the ten companies don’t include any information about leadership, suggesting that these companies aim to operate in a way that is detached from human knowledge, experience, and needs.  

What does this all mean? Well, it means that the vision for AI in agricultural technology (at least in the above ten companies), is being led by men. And in most cases, it is being led by white men from Western countries. Consequently, agricultural AI technology’s vision fails to actively integrate women’s perspectives in their direction and aims. Women farmers, who are responsible for half of the world’s food, are not actively or explicitly being considered in AI developments for agriculture. They are not even being advertised as key targeted users, even though women farmers produce about half of the world’s food.   

The failure of agricultural AI technology companies to include women and/or to be gender responsive is not unique. The AI industry at large (read PDF) is known to have discriminatory practices in the workplace that spread to its tools. Exclusive thought leadership produces exclusive products—products that worsen existing inequities, including gender inequities.  

To ensure that agricultural AI doesn’t make gender inequities in farming worse, there are three things that can be done:   

1. Companies can actively think about who is leading their process. They can ask, for example, “Where are the women?” and “How are women included in our thought leadership?” Companies can then act to include women’s perspectives, resulting in AI that is less discriminatory. Additionally, AI companies need to work to actively retain women and other individuals from underrepresented groups in AI. They should be transparent about hiring practices and provide regular trainings on equity, diversity, and inclusion to all employees.  
2. Companies can seek out women farmers as users of their products, using their guidance and perspectives on the technology and how it could help/hurt their agricultural practices. They can also represent these users on their websites, demonstrating their inclusivity and better representing the world’s farmers at large.  
3. Companies and researchers can work together to examine what the gendered consequences of AI will be more broadly. Important questions to consider include: How will decreased demand for labor impact men and women differently? What can be done to mitigate extreme financial losses for men and women? What policies need to be enacted to ensure that AI technology is distributed equitably, mitigating a potential “winners vs. losers" scenario in which women farmers are the likely losers given existing resource inequities?  

Women are key stakeholders in agriculture. Women account for 60 to 80 percent of smallholder farmers, producing 90 percent of food in Africa and approximately 50 percent of food worldwide. This means that women are key agricultural actors whose needs and perspectives must be considered to generate policies and technologies that benefit farmers and, ultimately, food security for everyone. Agricultural AI has the potential to help rectify existing inequities. To do this, though, AI leadership must be more inclusive and be more intentionally gender responsive in their technologies and goals.