Does this mean the end of AI use for writing workplace policies?
Employers across Australia have been advised to exercise caution when using artificial intelligence (AI) tools to draft workplace policies amid the advancing integration of the technology in the HR function.
BrightHR Australia, a software provider for workforce management and compliance, warned employers that using AI to create workplace policies can expose them to serious legal and compliance risks.
"AI can produce a workplace policy in seconds, but that does not mean the policy is legally compliant in Australia, appropriate for the business, or safe to rely on," said Mollie Eckersley, associate director of operations, Australia and New Zealand, BrightHR Australia.
"Employment law is highly dependent on which workplace relations systems it is governed by, which state the business operates in, and any specific awards or agreements that apply to the business or its employees."
Eckersley made the remarks after BrightHR's internal analysis identified critical gaps in AI-written absence management policies, including poorly crafted employee notification and return-to-work procedures.
The analysis also found inadequate recognition of statutory employee rights and limited guidance on how managers should handle individual circumstances.
Should HR stop using AI for workplace policies?
Australian HR leaders are significantly more likely to report advanced adoption of AI tools in their function than their counterparts elsewhere in the world, according to a recent Rippling white paper.
The paper, which polled more than 1,000 HR professionals globally in March 2026, found that 35% of organisations are already using AI tools for policy or documentation generation.
But the issue is not whether HR should use AI for workplace policies, according to BrightHR, which urged firms to get their workplace policies professionally reviewed and supported by practical guidance.
According to BrightHR, workplace policies affecting employee rights, performance, leave, or termination should be reviewed against current local workplace relations law.
"The issue is not whether businesses should use AI, but whether they understand where AI stops being a productivity tool and starts becoming a legal risk," Eckersley said.
"A policy that takes five minutes to generate could ultimately cost a business significantly more if it contributes to a dispute or an unlawful workplace decision."
