With all the focus on the EU AI Act and the United States, one nation has released a discussion paper on safe and responsible AI. The Australian government released its interim response on January 17 after conducting public consultations between June and August 2023. Submissions highlighted the opportunities of AI to improve wellbeing and grow the economy, but also concerns about potential harms. Views differed on the appropriate policy response.
Key themes in submissions:
Opportunities: AI can create jobs, benefit consumers, transform healthcare, and support the transition to net zero emissions. It also has the potential to improve education through personalized learning.
Risks: Technical limitations can lead to inaccurate or biased outcomes. A lack of transparency makes it difficult to predict errors, ensure accountability, and explain outcomes. Interactions with existing harms or laws may create new risks like online abuse or discrimination. Powerful new AI models could lead to large-scale, rapid harms. The pace of change also risks unforeseen consequences.
Policy response: Most called for government action to prevent and respond to harms, especially irreversible ones. But many applications are low risk and shouldn’t face burdensome rules. Suggested approaches include strengthening existing laws or new legislation focused on testing, transparency and accountability for high-risk AI. Non-regulatory initiatives like standards and capability building are also important.
The government’s interim response accepts current laws likely don’t adequately address AI’s risks. It commits to considering mandatory safety requirements for legitimate but high-risk AI uses where harms are irreversible. This could be through existing or new laws developed transparently with industry. Immediate steps include an AI safety standard, expert advisory group, and assessing labeling of AI content.
As for what happens next:
- Preventing harms through testing, transparency and accountability rules for high-risk AI.
- Clarifying and strengthening existing laws on issues like privacy.
- Advancing international collaboration, including the Bletchley Declaration on AI safety.
- Maximizing the benefits of AI through further investments in adoption and capability.
The response aims to ensure high-risk deployment of AI is safe and reliable while enabling low-risk innovation. It adopts principles like taking a risk-based and community-focused approach. Outside of the response, the Australian Government is keeping an eye on responses to AI around the world, including the EU AI Act.
If you’re wondering how Australia’s interim response, as well as other AI laws around the world, could impact you, reach out to BABL AI. Their Audit Experts are ready to help you with your concerns and questions while providing valuable assistance.
