Artificial Intelligence (AI) Tools
Examples of AI tools and categories and how they might apply to sports organisations.
In sport, typical AI tools include: 1
- Machine Learning (ML): Tools that are trained on large amount of data and continue to improve the more data they are trained on. Examples are systems that can predict sport injuries based on training on past injury data.
- Computer Vision: Technology that analyses images or videos, for example, to track techniques or assess skills.
- Natural Language Processing: Tools that work with human language, like translation services or chatbots for customer engagement.
- Generative AI: Systems that create new content, such as personalised training plans, automated match commentary, or even synthetic video.
- Agentic AI: Emerging technology that may, in future, be able to make autonomous decisions and act independently to achieve complex objectives.
Overview of AI categories, tools, uses, benefits, and risks in sport 1
Category | Example Tools | Sport Use Cases | Who Benefits Most? | Risks to Manage |
|---|---|---|---|---|
General-purpose AI | Google Translate, Grammarly, OCR apps, chat widgets (e.g., TourneyBot) | Automating admin tasks, improving communication, providing accessibility support | Volunteers, community clubs, CALD programs, inclusion initiatives | Not sport-specific (incorrect assumptions); privacy risks if data shared via public APIs; hidden biases; digital divide |
Generative AI | ChatGPT, Copilot, DALL·E, Gemini, Midjourney | Drafting newsletters, creating visuals, session planning, match summaries | Coaches, clubs, participants, marketing & comms volunteers | Inaccurate outputs; misuse of likeness/branding (deepfakes); copyright/IP issues |
Sport-specific AI | Zone7, STATSports, Playermaker, Hawk-Eye, Second Spectrum | Injury prediction, training load monitoring, officiating accuracy, tactical scouting | Elite and high-performance teams, federations, selectors | Over-reliance on black box models; athlete privacy and consent; limited access for grassroots clubs due to cost |
Comparison of AI Solutions for Evidence Synthesis in High-Performance Sport
In 2025, an Australian Sports Commission project evaluated the suitability of various AI-powered tools for evidence synthesis within sport science. Nine tools were assessed: five widely used, general-purpose large language models (in their free version) and four academically oriented or sport science–specific platforms (most requiring a subscription). Each tool was presented with identical sport science questions, and Subject Matter Experts independently performed blind reviews of all responses. The report concluded: 3
- For quick, general information needs, generic AI models may be appropriate, especially when users provide detailed, specific queries.
- For higher-quality evidence synthesis, academic-oriented tools are recommended.
- Users remain accountable for any advice or decisions informed by AI tools, and they must be able to explain, justify and take ownership of their advice and decisions.
