Precautionary Principle
A governance and decision-making principle that advocates implementing protective measures against potential AI harms even when causal relationships are not fully established scientifically or evidence is incomplete.
In plain language
When you are uncertain whether an AI might cause harm but the potential harm is serious, take protective action now rather than waiting for absolute proof. Better to be cautious upfront than to clean up a disaster later.
Why this matters
The precautionary principle is a sound risk management stance for AI governance, especially in high-stakes domains. Australian regulators expect organisations to demonstrate proactive risk management; waiting for certainty before acting is often indefensible when the consequences of failure are severe.
Relevance
StrategyUnderpins a proactive risk management governance stance that prioritises prevention over remediation, especially where AI harms could be significant.
Related terms
Putting precautionary principle into practice in your organisation?
Ready to transform your AI strategy?
Partner with Australia's AI strategy and governance specialists. From adoption roadmaps to ISO 42001 audit readiness.