As AI capabilities expand, concerns about privacy, bias, inequality, safety and security are growing. With this rapid advancement, it can be overwhelming for organisations to ensure their data remains safe.
The most highly publicised issue is queries and content submitted to AI engines unintentionally entering the public domain. Typically, this happens because the engine ‘learns’ from what is submitted to it, which means the data may be absorbed into responses to other users’ queries.
AI engines have to learn from something, and users’ data is a key source. Typically, ‘free’ AI engines learn from users’ data. As a general rule, if you don’t want something freely available on the internet, don’t upload it to an engine. Alternatively, check the small print and/or manage privacy settings to ensure your queries and content are not ‘training’ the engine. Typically, paid subscriptions to AI engines do not learn from user content, but do not assume this—always check.
Here are some other suggestions for helping ensure safe AI use:
- Select AI tools that have been thoroughly vetted for security. Begin by reviewing the security certifications and compliance standards of the tools you’re considering. Look for compliance with industry regulations like ISO/IEC 27001, or the world’s first AI management system standard ISO/IEC 42001.
- In New Zealand we also have the Privacy Act 2020 and the guidance on artificial intelligence and the IPPs. This guides organisations on how personal information is collected, stored, and used, ensuring the organisation handles personal data responsibly and securely.
- Manage data privacy settings. Most AI tools have features that allow users to disable data storage and it is key to make sure employees know when and how to navigate this. For example, OpenAI deletes chats within 30 days; however, their usage policy specifies that some chats can be retained for security or legal reasons. Consider having employees regularly delete chats using the AI tool’s settings to manage or delete chat history.
- Create a clear, simple AI use policy and ensure all staff understand your organisation’s expectations.
- Invest in regular staff training on AI best practices. When employees are aware of the risks and have clear guidance on organisational policies and expectations, they are better equipped to navigate potential challenges.
The AI Forum New Zealand is a rich resource of information, and you can also visit the associated Forum’s AI Governance site for toolkits, workshops and articles.