Thomas Lyons 
The Department of Agriculture, Food and the Marine (DAFM) has been employing artificial intelligence (AI) for several purposes, including predicting the likelihood of tuberculosis (TB) outbreaks. Fine Gael TD Ciaran Cannon has scrutinized the use of AI and the ChatGPT chatbot across government departments, including the DAFM. Cannon has pointed out “glaring inconsistencies” in the various approaches to using AI by government departments. DAFM is among the departments that have confirmed the use of ChatGPT. Cannon has said that there is a “complete inconsistency” in regulating and monitoring the platform’s use.
DAFM Minister for Agriculture, Food and the Marine, Charlie McConalogue, stated that the tool was “briefly explored” for answering technical or software-related questions when it was first launched. However, the minister emphasized that ChatGPT was not used to conduct business and that no official meetings were held regarding the application’s use. Minister McConalogue said, “It was determined that the tool was of little benefit and was not subsequently used. No data or information was uploaded or shared with the platform.”
The DAFM uses AI for several use cases, all of which are subject to human review, data protection, and governance measures. These use cases include predicting the likelihood of TB outbreaks, image analysis of crops claimed under Common Agricultural Policy (CAP) schemes, image analysis for identifying species susceptible to H5N1 (bird flu), customer segmentation analysis for developing agri-food policies, analysis for identifying risk factors for microbial food safety, and smart text analysis to prevent and contain data breaches. Additionally, the Area Monitoring System (AMS) within the DAFM uses machine learning, neural networks, and deep learning to process sentinel satellite data as required by EU regulations. The DAFM has contacted around 8,000 farmers after detecting potential agri-scheme application issues on their land parcels through the satellite inspection system.
The DAFM adopts a risk-based approach to the installation of any application used, and all applications must be used in line with DAFM policies and advice provided by the National Cyber Security Centre (NCSC), according to the minister. However, Deputy Cannon is calling for a complete ban on the use of ChatGPT across all government departments until the NCSC has assessed the implications of using the platform and issued the appropriate expert guidance. “I believe that there may be national security issues arising from the use of ChatGPT, particularly in the inputting of departmental data to produce the desired outputs from the platform. Right now we have no idea where that data, some of which could be highly sensitive, is ending up and who exactly has access to it,” Deputy Cannon said.
The use of AI in government departments has been a topic of discussion for some time. While the technology can provide significant benefits, such as increased efficiency and accuracy, there are also concerns about data privacy and security. As such, it is essential that government departments take a cautious approach to the use of AI and ensure that appropriate measures are in place to protect sensitive data. The DAFM’s risk-based approach is a step in the right direction, but it is crucial that all government departments take similar measures and work together to ensure that the use of AI is safe and secure.
In conclusion, AI is being used in several ways by the DAFM, including predicting the likelihood of TB outbreaks, image analysis of crops claimed under CAP schemes, and customer segmentation analysis for developing agri-food policies. While there are concerns about the use of AI, it is essential that government departments take a cautious approach and ensure that appropriate measures are in place to protect sensitive data. The DAFM’s risk-based approach is a positive step, and it is crucial that all government departments work together to ensure that the use of AI is safe and secure.
