Liming Zhu

The Australian Government has just unveiled a significant step in regulating AI, proposing 10 mandatory guardrails for high-risk AI applications. These guardrails aim to ensure testing, transparency, and accountability, crucial for managing the potential risks associated with AI systems, particularly in areas like critical infrastructure, medical devices, and biometric identification. This move marks a pivotal moment in aligning Australia's AI regulation with international standards, particularly those in Canada, the US, and the EU. It underscores the government's commitment to balancing innovation with safety and responsibility. As these regulations evolve, companies will need to be vigilant about compliance, especially in high-risk settings, to mitigate regulatory and operational risks and maintain stakeholder trust. Estimated reading time: 5 minutes https://buff.ly/47l2QH8 Need some help with AI, Tech or People? Visit our website (https://buff.ly/3ToD6UK) to schedule a call. #AIRegulation #ResponsibleAI #AustralianGovernment #TechCompliance #InnovationAndSafety #DigitalGovernance

1

🇦🇺 𝐓𝐡𝐞 𝐀𝐮𝐬𝐭𝐫𝐚𝐥𝐢𝐚𝐧 𝐠𝐨𝐯𝐞𝐫𝐧𝐦𝐞𝐧𝐭 𝐡𝐚𝐬 𝐥𝐚𝐮𝐧𝐜𝐡𝐞𝐝 𝐚 𝐧𝐞𝐰 𝐩𝐨𝐥𝐢𝐜𝐲 𝐟𝐨𝐫 𝐫𝐞𝐬𝐩𝐨𝐧𝐬𝐢𝐛𝐥𝐞 𝐀𝐈 𝐮𝐬𝐞 𝐢𝐧 𝐭𝐡𝐞 𝐩𝐮𝐛𝐥𝐢𝐜 𝐬𝐞𝐜𝐭𝐨𝐫, aimed at enhancing public trust and ensuring safe, ethical AI deployment. This policy, which will come into effect on September 1, 2024, sets a standard for AI transparency, governance, and risk management across all governmental agencies in Australia. 📝 The policy introduces mandatory guidelines and a framework for continuous adaptation as AI technology evolves. It emphasizes the importance of clear communication, public accountability, and training for all staff to align with the government's commitment to responsible AI use. #artificialintelligence #ai #ml #governance #aigovernance #responsibleai #grc #compliance #aicompliance #airegulation #law #regulation #machinelearning #technology #breezeml #aisafety #safety #policy #aipolicy #risk #riskmanagement #ethics #AIethics #GenAI #GenerativeAI #Australia #government

6

The Academy welcomes the release of the Australian Government’s proposals paper for introducing mandatory guardrails for AI in high-risk settings, and a voluntary AI safety standard. The documents released today are an important next step in the Australian Government’s Supporting Responsible AI consultation. (https://lnkd.in/eTxnxUTK) Read our response: https://lnkd.in/gMmkcGkx #AI #ArtificialIntelligence

12

As AI professionals, it's our responsibility to stay informed about these risks and work towards mitigating them.   MIT has created the world's most comprehensive AI Risk Repository, which is particularly useful for Australian businesses.   This repository consists of: - The AI Risk Database: Captures over 700 risks extracted from 43 existing frameworks, complete with quotes and page numbers. - The Causal Taxonomy of AI Risks: Classifies how, when, and why these risks occur. - The Domain Taxonomy of AI Risks: Categorizes these risks into seven domains and 23 subdomains.   These domains are: 1. Discrimination & Toxicity 2. Privacy & Security 3. Misinformation 4. Malicious Actors & Misuse 5. Human-Computer Interaction 6. Socioeconomic & Environmental Harms 7. AI System Safety, Failures, and Limitations   My takeaways Entity: 51% of risks are caused by AI systems themselves rather than 34% caused by a decision or an action made by humans. Timing: 65% of risks occur post-deployment (which emphasizes the importance of real-world testing and continuous risk assessment)   Top AI Risk Domains (based on papers and risks) 1. AI system safety, failures, & limitations (24% of risks) 2. Discrimination & Toxicity (16% of risks) - I find this particularly interesting and will delve into it this week.   Ps. For more information about the Australian Government's Eight Artificial Intelligence (AI) Ethics Principles, designed to ensure AI is safe, secure, and reliable, visit our website and interact with Alice, a Digital Human with the latest policy advice. Digital Human Assistants - https://lnkd.in/gRPDET6x. 

3