Atinar Pty Ltd

Action plan for making AI accountability a mandate across boards in UN member states Action plan for critical incident response protocols for cross-border AI incidents A global framework for AI covering safety, security, sovereignty, sustainability, supervision, scalability, standardisation, scalability and systemisation Action plan for non G20 countries to be protected from coordinated AI interference from AI powered nations Tabulation and Recognition of High Risk AI capability behaviour by massive superpowers including US and China Proposal for a non-proliferation treaty in AI beyond a set limit and governing AI research.

• Safe, secure and trustworthy AI
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Interoperability of governance approaches
• Transparency, accountability, and human oversight

There is an imminent need to recognise the widespread implications of self-regulated AI experiments. This has to undepinned by formulation of cross-border interoperable governance mechanisms which provides parity and balance to developing and non-tier 1 AI super power nations. This governance mechanism should intend to bring in necessary transparency and accountability in AI research, and deployment. Human supervision is critical in building safe and secure AI which is also sustainable, scalable. AI capability building and open source AI can be regulated once these frameworks are in place to ensure trustworthy AI which puts human ahead of everything else.

Clarity in policy and governance guidelines in counterintuitively stalling AI innovation. Organisations are unsure of upcoming policy changes and experimenting with AI in the fringes. This is resulting in productivity decline. Also, this trend is causing individual actors to bring their own AI into workplace and public service settings, resulting in potential leak of sensitive data and exposing critical services/data to AI induced cyber attack.

AI Dialogue can kick start the implementation of well defined action plans on various topics including AI governance and accountability mandates, cross-border AI incidents, AI impact on non-G20 nations, Safeguards for Bio and Chemical hazards induced by AI and

Formalise AI Governance as a mandate.

8 Action Committees focused on AI Systems (Capacity, Incidents, Tracking, Audit) , AI Standards (Governance, Accountability), AI Safety (Bio, Chemical, Nuclear), AI Security (Hacking, Autonomous Cyber Agents), AI Sustainability (ESG, Long Term impact on Pedagogy), AI Sovereignty (Data Residency, Cross-border protocols, Jurisdiction), AI Supervision (Human AI oversight, Human in the Loop), AI Scalability (Productivity in 3rd World, Scalability of Governance)

All risk averse voices are under-represented. The long list includes 3rd world nations, non-evangelistic AI policy makers, AI governance practitioners, AI safety champions, AI sovereignty supporters.

Our experience in facilitating such sessions, suggest that human centred design thinking techniques and workshops for ideating and converging on solution options will ensure that all voices get equal opportunity to articulate their point of view and democratic voting process are utilised to arrive at commonly grounded solution approach

NIST RMF - https://csrc.nist.gov/projects/risk-management/about-rmf ISO 23894 - https://www.iso.org/standard/77304.html ISO 42001 Atinar S8 Framework - https://zenodo.org/records/19412189 NAIC 6AI - https://www.industry.gov.au/publications/guidance-for-ai-adoption EU AI Act - https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai trustfor.ai - https://trustfor.ai Vanta