Australia has been urged to introduce stronger safeguards against the potential misuse of artificial intelligence, with experts warning that the technology could pose increasing risks to national security, public safety and workers.

Speaking during parliamentary hearings examining the safe introduction of AI, Australian Global Shield director Devon Whittle warned that foreign actors were already attempting to use advanced artificial intelligence to develop biological agents and weapons.

Whittle said Australia needed to improve its ability to respond to threats involving biological weapons, cyberattacks and AI systems operating in ways that were not intended or properly controlled.

He argued that governments should prepare for situations where AI systems fail or are deliberately misused, rather than relying solely on measures intended to prevent incidents from occurring.

The hearings also examined the growing use of artificial intelligence in Australian workplaces.

Shop, Distributive and Allied Employees' Association chief of staff Robert Tonkli raised concerns about employees being subjected to increased digital surveillance and automated workplace management systems.

He said AI was already being used to prepare staff rosters, allocate tasks, monitor performance and make decisions affecting employees, often with little or no consultation.

Tonkli described an incident involving a checkout worker with 30 years of experience who nearly lost their job after workplace software identified errors and automatically triggered a first-and-final warning.

He warned that automated systems did not always account for individual circumstances or the practical needs of workers, potentially resulting in unfair disciplinary action and additional workplace stress.

The union argued that artificial intelligence should also be treated as a workplace health and safety issue and recommended establishing a dedicated AI panel within the Fair Work Commission.

The evidence presented to the parliamentary hearings highlighted concerns about both the security risks of increasingly advanced AI systems and the consequences of using automated technology to make decisions affecting people's employment and working conditions.