Australia to adopt 'mandatory' AI safeguards

Australia plans to enforce mandatory AI safeguards for companies operating within its borders, Prime Minister Anthony Albanese announced on Tuesday. The move comes in response to growing global concerns over the potential risks associated with advanced AI technology. Following a series of hacking incidents involving prominent AI firms like OpenAI and Anthropic, Australia will establish a new AI safety regulator.

Albanese emphasized the importance of shaping AI to benefit society rather than being shaped by it. The measures would encompass mandatory reporting of AI security breaches and independent evaluations of frontier AI models. Australia, recognizing its strategic position for data centers supporting AI, is a pioneer in this area, aiming to leverage the benefits of AI while addressing social concerns.

However, the government acknowledges the need for robust risk management and independent oversight of AI models. Defence Minister Richard Marles stressed the importance of external scrutiny to ensure the safety of AI tools. With renewable energy resources and expansive uninhabited lands, Australia is well-positioned to support the data centers required to fuel AI advancements.

The government is also grappling with the contentious issue of copyright law amendments to allow chatbots to train on creative works.