What does AI think of NZ’s AI strategy?

AIGovernment
Written By Vic MacLennan

I asked 3 x generative AI tools the following question:

Compare https://www.mbie.govt.nz/assets/new-zealands-strategy-for-artificial-intelligence.pdf to other countries and rate New Zealands strategy as a percentage relative to leading countries

From worst to best their responses in a nutshell:

  • ChatGPT - 55%

  • CoPilot - 65%

  • Gemini AI - 80-85%

You’ll find their full responses below — but first, what does this range actually tell us?

Media coverage and much of the tech industry (with a few notable exceptions) have been fairly critical of New Zealand’s strategy to date. Two of the three tools reflected that sentiment. But the real insight here is the variation in responses. I asked all three tools the exact same question, in a clean slate session, without any prior context — and yet their assessments diverged wildly.

So, what’s going on?

Is there inherent bias in how these tools interpret national strategies? Was Gemini perhaps trained on more government-sourced or optimistic material? Or could it even reflect the voice of the author — was Gemini the tool used to help write the strategy?

Another possibility is that each model weights different factors:

  • ChatGPT might favour global regulatory benchmarks,

  • Copilot may lean toward pragmatic implementation,

  • Gemini could over-index on policy tone or intent.

Either way, it’s a fascinating reminder that even AI "opinions" are shaped by how they’re trained, what data they’re exposed to, and what priorities they assume. When evaluating national strategies — or anything else — it's worth remembering that generative AI doesn't just mirror reality, it interprets it.

Response from OpenAI ChatGPT

🇳🇿 New Zealand’s AI Strategy: “Investing with Confidence”

🔝 Global Benchmarks

🇺🇸 United States

🇪🇺 European Union

🇬🇧 United Kingdom

📊 New Zealand’s Relative Position

Overall55% A solid, pragmatic strategy—but falls short of superpower-style ambition.

💡 Takeaway

New Zealand’s AI Strategy is a pragmatic, adoption-first approach that empowers government and business through training, ethical alignment, and supportive policy. However, it lacks the binding regulation and ambitious investment seen in the EU, UK, and US.

If NZ aims to elevate its global AI standing, it could consider:

  • Establishing an AI safety oversight body, akin to UK/US models;

  • Investing in sovereign AI infrastructure or research capacity; and

  • Tightening responsible AI regulation beyond existing privacy and fairness frameworks.

At ~55%, NZ is on the right track—but still needs bold investment and capacity-building to truly compete with leading AI nations.

Response from Microsoft Co-Pilot

Based on the latest national AI strategies from leading countries and the comparative analysis from the AI and Democratic Values Index 2025 (AIDV-2025), here’s how New Zealand’s AI Strategy—“Investing with Confidence” (July 2025)—compares to global leaders:

 🔍 Summary of New Zealand’s AI Strategy

New Zealand’s strategy emphasizes:

  • AI adoption and application over foundational AI research.

  • Private sector enablement, especially SMEs.

  • Responsible AI guidance for businesses.

  • International collaboration and alignment with global standards.

  • Leveraging existing science and innovation systems.

It is pragmatic and tailored to New Zealand’s economic scale and capabilities, focusing on trustworthy, inclusive, and applied AI rather than competing in frontier AI development.

📊 Why New Zealand Scores ~65% Relative to Leaders

  • ✅ Strengths:

    • Clear focus on responsible AI and business enablement.

    • Strong alignment with international norms.

    • Realistic and context-aware strategy.

  • ❌ Limitations:

    • Lacks significant investment in foundational AI R&D.

    • No large-scale infrastructure (e.g., AI factories, national compute resources).

    • Less emphasis on global leadership or exportable AI innovation.

🧭 Final Thoughts

New Zealand’s strategy is well-calibrated for its size and economic context, prioritizing trust, adoption, and practical impact. While it doesn’t aim to lead in cutting-edge AI development, it positions the country to benefit from global AI advances responsibly and inclusively.

Response from Google Gemini AI

New Zealand has recently published its "Strategy for Artificial Intelligence: Investing with Confidence" in July 2025. This strategy is a significant step, as New Zealand was previously the only OECD country without a formal AI strategy. The core focus of New Zealand's approach is on practical AI adoption and application rather than competing in high-end foundational AI development. The government aims to foster an environment where businesses can confidently invest in AI, with a projected economic benefit of $76 billion to the New Zealand economy by 2038 from generative AI alone, representing over 15% of its GDP.

Key elements of New Zealand's AI Strategy include:

  • Government's Role as an Enabler: The strategy emphasizes reducing regulatory uncertainty, proactively removing barriers, and providing clear guidance for businesses to confidently adopt AI.

  • Adoption of OECD AI Principles: New Zealand explicitly aligns its approach with international best practices by adopting the OECD's AI Principles, committing to innovation, transparency, and human rights in AI development and deployment.

  • Addressing Identified Barriers: The strategy directly tackles common obstacles to AI uptake in New Zealand, such as regulatory uncertainty, perceived complexity and ethical concerns, lack of understanding, and skills shortages. Actions include publishing responsible AI guidance, increasing awareness of AI's benefits, and investing in workforce capacity building through education and training programs.

  • International Collaboration and Investment: New Zealand seeks to engage in global AI governance forums and attract foreign direct investment (FDI) in AI, exemplified by initiatives like the Singapore-New Zealand AI for Healthy Ageing Initiative and changes to investment policies.

  • Leveraging Existing Strengths: The strategy highlights New Zealand's Science, Innovation, and Technology (SI&T) system, with plans to create a new advanced technology public research organization and optimize government funding for AI-related research and development.

Comparison to Leading Countries' AI Strategies:

Leading AI countries generally include the United States, China, the United Kingdom, Canada, Singapore, and several European nations like France and Germany. While their strategies share common goals like economic growth, innovation, and responsible AI, they differ in their scale of investment and primary focus:

  • Investment: Countries like the United States and China are making massive investments in AI, with the US projected to invest over USD 470.9 billion in 2025, and China investing tens of billions through national funds. These nations are heavily focused on foundational AI research and development, aiming to be global leaders in creating cutting-edge AI technologies.

  • Foundational vs. Adoption Focus: Unlike the US and China, whose strategies heavily emphasize foundational AI research, New Zealand explicitly prioritizes adoption and application of AI to solve local challenges and leverage its competitive advantages in sectors like agriculture and healthcare.

  • Regulatory Approaches: While all leading countries are grappling with AI governance, approaches vary. For instance, the UK has initially opted against new legislation, tasking existing regulators to apply AI principles. The EU has a comprehensive, legally binding AI Act. New Zealand, similar to the OECD approach, focuses on a light-touch, principles-based regulation that updates existing frameworks.

  • Talent and Ecosystem: Leading countries invest significantly in developing a large pool of AI talent and fostering a robust ecosystem of AI companies and research institutions. New Zealand acknowledges a skills shortage and aims to build "Kiwi capacity" through specific educational and training initiatives.

Rating New Zealand's Strategy as a Percentage Relative to Leading Countries:

When evaluating New Zealand's AI strategy, it's crucial to consider its stated objectives and the scale of its economy. New Zealand is not attempting to outspend or out-develop AI giants like the US or China in foundational research. Instead, it aims to be a "leader in responsible AI adoption" and a "confident, capable, and responsible player in the global AI economy".

Given its pragmatic focus, clear identification of barriers, detailed action plans, commitment to ethical AI aligned with international standards (OECD Principles), and initiatives to attract investment and build domestic capacity, New Zealand's strategy is well-suited for its context.

The "Global AI Index 2024" already ranks New Zealand 8th globally for AI Preparedness, with a score of 0.75 out of 1. This indicates a strong position compared to other nations. While it's catching up in the rate of AI adoption among businesses compared to some countries like Australia, its strategy is forward-looking and comprehensive for a country of its size.

Therefore, relative to leading countries, and considering its strategic objectives and well-defined plan for a smaller, advanced economy, New Zealand's AI strategy can be rated as approximately 80-85% of what could be considered a leading-edge strategy. It demonstrates a clear understanding of the global AI landscape, its unique position, and a pragmatic roadmap for success.

___________________________________________

Quick reminder of our recent cartoon on the NZ AI strategy.

Vic MacLennan

CEO of IT Professionals, Te Pou Haungarau Ngaio, Vic believes everyone in Aotearoa New Zealand deserves an opportunity to reach their potential so as a technologist by trade she is dedicated to changing the face of the digital tech industry - to become more inclusive, where everyone has a place to belong. Vic is also on a quest to close the digital divide. Find out more about her mahi on LinkedIN.

Previous

ITP Cartoon by Jim - Promptductivity
Next

SharePoint zero-day hack hits hard