One popular alternative to the current political-economic landscape of food lies in bringing agriculture to the modern, technologically advanced age though big data. For agriculture, big data is just what it sounds like, collecting vast amounts of data from crop and weed locations, soil and weather conditions, and pesticide and herbicide use. There are several technologies to collect big data depending on the kind of data being collected including sensors in the soil and on plants to weather stations and even drone technology (Carbonell 2016). Once the data is collected, it can be used to both interpret past events and predict future events. Especially for predicting future events, this has huge implications for how farmers make decisions such as where to plant crops, when and how much to water crops, and the timing and quantity of pesticides and herbicides to use. Because of all this additional information, big data allows for more precise and efficient use of resources by farmers (Carbonell 2016).
However, on its own, big data is practically useless without a way to interpret it. The way most big data works is through the use of autonomous technology, or artificial intelligence (AI), described as “systems in which the designer has not predetermined the responses to every condition. Such systems are capable of surprising their designers and behaving in unexpected ways” (Omohundro 2014). This technology is attractive because of the ability to synthesize many complex variables in order infer relationships and make informed, split-second decisions. Because of this, AI is gaining huge popularity around the world for many different applications, including agriculture.
Together, big data and AI have the possibility of making huge changes in the food system. Yet, this is not the first time farming has ever collected data to inform its practices. For more than a century, farming has been empirically driven through tracking property and capital use and resources such as the Farmer’s Almanac (Bronson and Knezevic 2016). In fact, it can be argued that big data does not provide a big enough shift to be considered an alternative to the current food system. Yet, big data represents the first time this extent of information is collected and digitized for easy analysis. Also, beyond the vast extent of data being collected, we are also seeing the “digitization” of data that has never existed before. In agriculture is represented through new statistics of plant growth and more location specific data for soil and weather conditions (Cukier and Mayer-Schoenberger 2013). Because of these changes, I argue that there is a profound and important shift occurring in food systems due to this new technology.
Several large corporations and investors also believe big data to be the future of agriculture. For example, biotech giant Monsanto bought Climate Corp, one of the largest agriculture data companies, in 2013 for US$930 million (Carbonell 2016). This signals a large change in big data, making it a much more competitive and high-profile field. Another giant agribusiness to have invested in big data is John Deere, an agriculture and construction automobile company, that is now designing autonomous tractors that communicate via GPS in order to track real-time harvests and usage of fertilizers (Satariano and Bjerga 2016). Other large biotech and chemical companies such as Syngenta, DuPont, ChemChina, and Dow Chemical are also acquiring big data companies or otherwise incorporating big data into their business strategies (Satariano and Bjerga 2016). Overall, funding for agriculture technology startups has rapidly grown to US$4.6 billion in 2015, doubling the US$2.3 billion invested in 2014 (Satariano and Bjerga 2016). While only a part of this is going to data companies, investment for these businesses is growing enormously, signalling that big data is a growing field that is considered lucrative and attracting the attention of many smart entrepreneurs and investors.
Big data is already being deployed to varying degrees in farms across the world. Many farms, especially larger industrial farms, already use some sort of sensors in order to track basic variables such as soil water content, soil salinity, or microclimate (Bronson and Knezevic 2016). In 2010, 72% of corn acreage in the US was planted using precision technology, increased from 17% in 1997 (Carolan 2016). However, much fewer are using big data to the extent that is technologically available along with AI for data interpretation. Big data is largely targeted at large industrial farms, especially in the US, due to having more variables to keep track of and being more willing to test new technologies. In addition, most of these technologies are targeted at and most useful for single-crop farms (Carbonell 2016).
Some of the biggest common factors driving big data implementation in agriculture include anticipation for the future and moral judgements. Among interviews conducted by sociologist Michael Carolan with farmers, big data analysts, and food entrepreneurs, these were the factors that were present for all parties (Carolan 2016). The first reason, anticipation for the future, is mostly for the fear of the growing population and environmental disasters. As Mark, a farmer, succinctly puts it, “Two words: 9 billion. I’m doing this so we don’t all starve” (Carolan 2016). This shows that the biggest reason big data is so popular is because it represents a step towards making food more efficient and more productive. Secondly, moral judgements play a large role in the adoption of big data as well. Sonia, who works in the big data industry, writes, “It takes a lot of the guesswork out of farming. Farmers use precision technology to make better farm management decisions. It makes a good farmer better” (Carolan 2016). Prevalent among everyone interviewed was that this technology will make one a “better” farmer. Increasing yields can create a reputation and make a farmer look better to consumers, other farmers, or oneself.
While these were common factors in interviews, both of these factors are very debatable and make large assumptions. For instance, it makes sense that being able to grow more food with less resources will help to feed a growing population. However, this ignores dozens of other factors, including the question of distribution and equity once more food is being produced, and it is difficult to imagine big data greatly affecting the problems that currently plague global distribution of food. Also, the moral claim that big data will make one a “better” farmer is entirely culturally constructed. While being more resource efficient may translate to being a better farmer in some contexts, this is not likely to translate for every farm. Additionally, being more efficient may result in, but certainly does not necessitate more food being produced or higher profit margins for farmers.
Another common thread that Carolan noticed through his interviews was the amplification of “absent forces”. In other words, there was often a disconnect between the technology and the rest of the world. For example, Lisa, a regional food systems entrepreneur writes, “These are grassroots technologies; they’re a response to the needs of a community. We don’t need to convince people they need them – the typical pathways of most new innovations. So I’m not as interested in the product as in the question of whether I’m heading in a direction that removes barriers. This is a team effort, and I mean that in the broadest sense” (Carolan 2016). This statement assumes that a farmer would completely rely on the absent presence of the company providing while also assuming a sense of inevitability of the technology that may not be universally accepted.
The future of big data is still undetermined, but there is this sense of inevitability of the technology due to the numerous, quantifiable benefits it provides. Yet, there are plenty of reasons to be hesitant towards big data as well. For example, one of the biggest implications for big data is the redistribution of power in agriculture. With big data, farmers rely on either data companies or large agribusiness conglomerates in order to have access to and interpret their data (Carbonell 2016). This continues a trend of farmers having less agency in favor of corporations. In the words of Douglas Hackney, president of a data management business, “For a big data company, what is a farmer? It’s an account number… for a farmer, if their data falls into the wrong hands, it’s an existential threat” (Carbonell 2016). This threat of data being controlled by someone else could have huge effects for commodity market speculation and has potentially detrimental implications for crop prices. It is understandable that a farmer might have concerns over how their data could be abused by another party.
This is reminiscent of the recent situation with farmers relying more on corporations with the implementation of genetically modified seeds in favor of traditional seed saving practices. In both transgenic seeds and big data, individual farmers are paying corporations to produce crops better and more efficiently than the farmer could with only his or her own knowledge and skills (Mascarenhas and Busch 2006). The new farmer then does not need the skills and knowledge that previous generations of farmers required. Instead, the new farmer relies on corporations to a much greater extent. However, certain companies and projects are working on creating environments where farmers are in control of their own data. Some examples are ISOBlue, FarmLogs, and Open Ag Data Alliance, all helping farmers access and control their own data (Carbonell 2016).
In addition to corporations gaining more power, we are also seeing entrepreneurs becoming more influential with the popularity of big data (Carolan 2016). As discussed earlier, there are huge and rapidly growing investments currently going into AgTech. With this, food entrepreneurs have the power to better inject new and inventive technological ideas into big agriculture, often times being later bought by larger food corporations (Carolan 2016). In particular, there are hundreds of startups working on everything from unexplored crop-specific precision agriculture, better big data interfaces, and artificial intelligence built into farming infrastructure.
There are also many fears around big data and AI in agriculture. These fears are mostly founded in the fact that the technology is still so new and undeveloped. As outlined in anthropologist James Scott’s book, Seeing Like a State, one such fear is that a government power would abuse the abundant data without the appropriate knowledge of agriculture (Cukier and Mayer-Schoenberger 2013). It would not be difficult to imagine the data being exploited in order to meet a quota without concern for other factors such as labor or sustainability.
Another place where fear is apparent is in how the general consumers may react to the influx of big data and artificial intelligence in the food industry. Take, for instance, this package of soba noodles that advertise “No artificial intelligence used in the making of Organic Planet!”.
This unique label could be referring to some of the equity issues associated with artificial intelligence as discussed earlier. A consumer may choose to purchase Organic Planet in order to protect farmers from reliance on corporations and to protect their rights to their data. However, while there may be intentions of equity, since the package makes no reference to farm labor or equity, it appears that this label is more of an appeal to the technology-adverse population. There appears to be a growing distress around AI that may be because of a fear of unknown or unintended consequences as well as the Luddite fear of losing jobs to AI or robotics. Big data and especially artificial intelligence could easily become villainized through rhetoric in advertisements and labels such as this one. Whether or not this fear could be an actual threat to the market remains to be seen, but it is important to note that there is already limited market pushback for big data and AI.
Knowing these benefits and drawbacks, the question remains: does big data make for an adequate alternative to the current industrial food system? This is an especially important question if the adoption to big data will be as inevitable and widespread as those in the industry believe. With all of the benefits of the technology, it is clear that large improvements in efficiency and financial sustainability could be made through big data. However, there are all clear equity concerns as well. One way to address these concerns is through farmers privately owning all of their data. Alternatively, data could be open-sourced and publically available, but anonymous in order to prevent exploitation (Carbonell 2016).
In addition, big data will never make an ideal food system alternative if it does not begin to address how it can be applied to different kinds of agriculture. This technology has interesting applications for farms that are not the large, monocropping, technology apt farms that are currently being targeted. Assuming big data continues to grow in popularity, it will become necessary to look at how it can benefit smaller, ecologically diverse, and culturally different farms. I would recommend careful studies that look into the implications of how big data is currently being used and further studies as new and improved technologies are tested on farms. Further technological and policy recommendations can be made based on these studies in order to ensure a more equitable food system with big data.
In conclusion, big data, along with AI, presents an interesting and important opportunity for the future of food systems that deserves careful attention. The funding and support for the technology is growing greatly, and if implemented with care, big data may be a powerful technology that will make growing crops easier and more efficient than ever before. If not careful though, it may also further the power divides in big agriculture and lead to the further exploitation of farm labor. Additionally, there are many still unknown questions left for big data. For example, even if implemented equitably, could big data have any effect on current inequalities in food distribution? In what ways can big data be implemented in different contexts? And how are consumers going to react the products? It is clear that more research needs to be conducted in order to grasp the full implications of big data on our food systems.
Works Cited
Bronson, Kelly, and Irena Knezevic. 2016. “Big Data in Food and Agriculture.” Big Data & Society 3 (1): 2053951716648174. doi:10.1177/2053951716648174.
Carbonell, Isabelle M. 2016. “The Ethics of Big Data in Big Agriculture.” Internet Policy Review 5 (1). doi:10.14763/2016.1.405.
Carolan, Michael. 2016. “Publicising Food: Big Data, Precision Agriculture, and Co-Experimental Techniques of Addition.” Sociologia Ruralis, January. doi:10.1111/soru.12120.
Cukier, Kenneth, and Viktor Mayer-Schoenberger. 2013. “The Rise of Big Data: How It’s Changing the Way We Think about the World.” Foreign Affairs 92: 28. doi:10.2469/dig.v43.n4.65.
Mascarenhas, Michael, and Lawrence Busch. 2006. “Seeds of Change: Intellectual Property Rights, Genetically Modified Soybeans and Seed Saving in the United States.” Sociologia Ruralis 46 (2): 122–38. doi:10.1111/j.1467-9523.2006.00406.x.
Omohundro, Steve. 2014. “Autonomous Technology and the Greater Human Good.” Journal of Experimental & Theoretical Artificial Intelligence 26 (3): 303–15. doi:10.1080/0952813X.2014.895111.
Satariano, Adam, and Alan Bjerga. 2016. “The Weather-Predicting Tech Behind $62 Billion Monsanto Bid.” Bloomberg Businessweek, June 9.
Sonka, Steve. 2014. “Big Data and the Ag Sector: More than Lots of Numbers.” International Food and Agribusiness Management Review 17 (1).