Artificial intelligence is revolutionizing the sports industry, and sports betting is no exception. Organizations and applications are increasingly leveraging AI to enhance the entire betting ecosystem. From refining betting odds to predicting outcomes and detecting fraud, AI is reshaping how we engage with sports betting. In this article, we’ll delve into the intersection of AI and sports betting, exploring how cutting-edge technology can elevate your strategies and provide valuable insights and offering practical tips for mitigating risk and navigating the complex legal landscape.
Uses of AI in sports betting
At the center of AI-based sports betting are AI algorithms, sports data -- including game scores, statistics, and athlete health and sensor data -- and data analytics. AI algorithms can analyze massive amounts of sports data to make predictions and recommendations. They can also analyze historical data, player performance and contextual factors to predict and match outcomes. These AI-driven models can assist bettors with making informed decisions, potentially improving their chances of winning.
The speed and power of the latest generation of AI technology are also used to dynamically adjust betting odds based on real-time information. By continuously recalibrating odds, bookmakers can ensure they accurately reflect the current state of play. This real-time information can also be used in connection with AI-driven trading bots that execute bets automatically. These bots can capitalize on market fluctuations, identify arbitrage opportunities, and execute trades quickly, thereby enhancing efficiency and responsiveness in sports betting.
Privacy and intellectual property considerations for sports data and AI algorithms
As is the case with many other applications of AI, the use of AI in sports betting comes with both opportunities and potentially significant legal risks. One of the biggest risks related to the use of proprietary and personal data in the AI models is determining which data may be protected by legal rights or subject to written contracts or terms of use. The proliferation of privacy laws in the United States has been remarkable. From the EU General Data Protection Regulation (GDPR) to the California Consumer Privacy Act (CCPA), regulators worldwide are emphasizing the protection of consumer privacy. These comprehensive privacy laws often define broad categories of protected information, encompassing both personally identifiable data and non-identifiable data. As businesses increasingly rely on AI and machine learning to process personal data, they must navigate legal obligations related to data privacy. When it comes to proprietary data that does not constitute personal data, it is important to understand the terms and conditions under which the data is made available. For example, the reporting of some sports scores is restricted by the terms and conditions of the data provider (Morris Communications Co. v. PGA Tour, 364 F.3d 1288, 11th Cir. 2004) and many sports metrics providers also define terms of use of their data on their web pages.
When utilizing sports data and AI algorithms, intellectual property rights become another critical area to navigate. Depending on the nature and context of the AI (Ex Parte Hannun, Appeal No. 2018-003323, US Patent Trial and Appeal Board, Apr. 1, 2019), the methods used for training and inferencing the AI algorithms may be eligible for protection under intellectual property laws. For instance, innovative processes or techniques that leverage AI to generate real-time sports betting odds or similar recommendations could potentially be patentable (Ex Parte Setty, Appeal No. 2021-000561 (US Patent Trial and Appeal Board, June 15, 2021). To mitigate risks, it is important to assess the possibility of infringing third-party intellectual property rights before developing or deploying any AI technology.
Additionally, ownership of the underlying algorithms must be carefully considered. If open-source software was employed during the development of the algorithms -- a common practice among developers -- the entire algorithm could be subject to the terms of the associated open-source license. These terms might include restrictions on charging license fees for software developed using open-source code or obligations to make all modified source code available to the public. In summary, having a well-defined open-source policy and monitoring software usage within algorithms are crucial steps in managing intellectual property considerations.
Strategies for mitigating risks in sports data and AI
Copyright protection
Copyright protection plays a crucial role in safeguarding intellectual property related to sports data. While data itself is generally not patentable (information is not considered one of the four statutory categories of patent-eligible subject matter under 35 U.S.C. § 101), copyright offers a potential avenue for protecting sports data. Historically, live sports game statistics have not been enforceable in courts as intellectual property. However, there remains a risk of copyright infringement claims. To mitigate this risk, consider transforming sports data used in AI algorithms or utilizing small subsets of the data.
Legal considerations when using sports data
When acquiring sports data for AI algorithms, it’s essential to navigate legal considerations. Whether you’re downloading, scraping, or inputting sports data from a website or application, ensure compliance with contracts and terms of use set by the data provider. Additionally, be aware of any restrictions imposed by sports data providers to protect their data.
Patent applications and AI-related Inventions
The patentability of AI-related inventions depends significantly on the level of detail and description provided in patent applications. The US Patent Office has allowed AI-related patents at varying rates based on interpretation. However, disclosing AI-related sports betting data in a patent application carries some risk. Once public, this information becomes accessible to competitors. Balancing the benefits of protecting intellectual property against the potential risks of disclosure is essential.
Ownership agreements
When contracting software developers to create algorithms, establish clear ownership agreements. Ensure that any intellectual property rights associated with the algorithms are assigned to the company. This proactive step minimizes the risk of ownership disputes down the line. By following these strategies, organizations can navigate the complex landscape of sports data and AI while safeguarding their intellectual property interests.
In conclusion, the integration of AI into sports betting presents both remarkable opportunities and notable legal challenges. As AI algorithms become increasingly sophisticated in analyzing sports data and predicting outcomes, it is crucial for data providers and users to navigate the complex legal landscape carefully. Protecting personal information, respecting terms of use and ensuring proper ownership and licensing of AI algorithms are essential steps to mitigate risks. By addressing these considerations, stakeholders can harness the power of AI to innovate and monetize the technology within the sports betting industry while safeguarding against potential legal disputes and maintaining the integrity of personal data.
Baird Fogel is the U.S. Sports Practice lead and partner-in-charge in Eversheds Sutherland’s San Francisco office. Rachel Reid is the U.S. head of artificial intelligence and partner at Eversheds Sutherland. Joshua Branson is counsel in the firm's Intellectual Property Department. Also contributing to this column were Joel Bradley, a senior associate attorney; and Kristi Kimiko Thielen, an associate attorney.
