Artificial intelligence (AI) is quickly becoming part of every-day decision making on many farms across Australia.
Wearable collars, ear tags, cameras and sensors can monitor animal movement and eating habits, allowing AI systems to detect and generate recommendations that would previously have required great amounts of manual labour and surveillance.
In the cattle industry, platforms such as Beef AI are helping producers analyse large amounts of data on animal genetics and performance to support more informed breeding decisions. Other tools like Ruminants AI use predictive modelling and image analysis to recommend diets and identify nutritional issues before they become serious.
The opportunities are endless. AI can assist producers to improve breeding outcomes, identify illnesses earlier, predict calving dates and overall make more informed operational decisions.
However, as AI becomes increasingly embedded in agricultural businesses, there are a number of legal and commercial questions that arise.
Agricultural data
Agricultural AI platforms rely on large volumes of information from farms, including, breeding records, herd performance data, animals’ health information and pasture and environmental data.
The more data an AI platform can access, the more effectivity it can identify patterns and improve recommendations.
Many farmers will assume that because the information originates from their own property and operations, they own it. More importantly, who has the right to use that data.
Like many AI platforms, users are required to agree to detailed terms and conditions before accessing the service. These agreements may provide the platform with rights to collect, store, analyse and use the data. In some cases, AI providers may also have rights to share or commercialise any information provided by users.
Who owns AI-generated recommendations?
Agricultural AI systems may generate recommendations relating to breeding decisions, feed rations, grazing management, and animal health interventions. These recommendations are derived from a combination of farm-supplied and large datasets contributed by users. This raises another important question: who owns the resulting output?
Is it the producer whose data contributed to the recommendation? The tech provider that developed the AI model? Or both?
Australian intellectual property laws do not provide a straightforward answer to every scenario involving AI-generated material. In many cases, the rights to AI-generated outputs will depend not only on intellectual property legislation, but also on the contractual arrangements governing the platform.
Producers should therefore consider not only what data they are providing to an AI platform, but also what rights they receive in return and what rights the provider retains over the outputs generated by the system.
Are AI-generated recommendations protected by copyright?
Under Australian legislation, copyright generally requires the material to be created by a human. Where recommendations are generated primarily by an AI system with limited human involvement, there may be uncertainty as to whether copyright protection exists at all.
Even where copyright protection is available, the platform’s terms and conditions may determine who owns the resulting material and how it may be used.
As AI technology continues to develop, intellectual property law is likely to face increasing pressure to address questions surrounding authorship, ownership and control of AI-generated material more directly.
What happens if AI gets it wrong?
There are great benefits to AI, however, it is no substitute for human experience and expertise.
AI is not immune from error. Poor quality data and biased algorithms can result in flawed recommendations.
This raises an important legal question: who bears responsibility when an AI-generated recommendation causes loss?
Unlike humans and corporations, AI systems are not recognised as valid legal persons under Australian law. Responsibility therefore falls on individuals and organisations involved in developing, supplying or managing the platform.
Depending on the circumstances, potential liability may arise under areas of law including negligence, contract and consumer or product liability. However, the application of existing legal principles to AI-related disputes is still developing.
A recurring theme across the agricultural AI sector is that AI can assist decision-making, but it cannot replace human judgment.
AI systems are capable of identifying patterns across datasets, from animal health records to weather forecasts. However, they can produce inaccurate outputs.
This creates a practical dilemma for producers. What happens when an AI recommendation conflicts with a farmer’s own experience and judgment? If the producer follows the recommendation and there is loss, responsibility is not straightforward.
Until liability and accountability frameworks become clearer, producers should view AI as a tool to support their decision-making rather than a substitute for their own judgement.
What Should Agricultural Businesses Consider?
AI presents significant opportunities for Australian farmers, but it also raises important legal and commercial risks.
Before adopting an AI platform, agricultural businesses should consider:
- What data is being collected, who owns it and how can it be used?
- Who owns AI-generated recommendations and other outputs?
- How is commercially sensitive information protected?
- What happens if an AI recommendation is wrong and causes loss?
- Do the platform’s terms adequately protect your data, confidential information and intellectual property, and clearly limit how the provider can use and share that information?
Our IP + Technology team can assist you with AI platform terms and agreements. Please contact Ersel: ersel@morganenglish.com.au
