NZ ranks as 2nd most 'AI-nervous' country

PLUS, major Kiwi health groups launch joint AI venture

Author

Mike Bayly
August 04, 2025

Kia ora! Welcome to New Zealand’s weekly roundup of AI news, events, jobs and education.

Thanks to everyone who completed our AI event survey. Your feedback made it clear where the demand is, so stay tuned for the event announcement coming soon. We are also shaping our content around what you told us, so expect changes in Mike’s LinkedIn content and in our newsletter’s Levelling Up section.

Now, this week’s viral story: ChatGPT outperformed investing market benchmarks by 24% in just one month while managing a live portfolio, offering a glimpse of where AI-driven investing could be headed.

This week’s highlights

😨 NZ ranks 2nd most AI-nervous
🧠 Uni tool predicts dementia early
🧳 AI trip planner boosts Kiwi tourism
🗣️ AI speech coach aids te reo revival
🌐 US, China pitch global AI gameplans
🖥️ Datacom goes all-in on Kiwi-made AI
🏥 NZ health giants launch AI joint venture
🎙️ Karrina Mountfort: Amplifying women in AI

Happy reading and listening ✌️

Did someone forward you this? Sign up!

HEADLINE STORY

📊 Kiwis Second Most AI-Nervous Globally Despite Growing Understanding

NZ Results from the IPSOS AI Monitor 2025 Survey.

📣 Word On The Street: Two-thirds of us get nervous about AI products despite 73% now understanding the tech well.

🔍 Zooming In:

  • 66% of Kiwis nervous about AI: Second highest globally, just behind Australia (67%)

  • 73% understand AI well: Massive jump from 62% in 2023 and 69% in 2024

  • Only 39% trust companies to protect personal data when using AI (vs 79% global average)

  • 84% prefer human-created content for news articles and photojournalism over AI

  • 58% believe AI will worsen disinformation: Places us 3rd highest for this concern globally

🏘️ Our Take: The trust gap matters for NZ businesses rolling out AI tools: customers want transparency about when AI is being used. Higher AI literacy (73% vs 62% in 2023) shows we're getting smarter about the tech, which could drive better adoption if privacy concerns are addressed. Kiwis needs targeted education that turns skepticism into strategic advantage:

  • Business leaders require transparency training: learning how to communicate AI use clearly and build customer trust through proactive disclosure rather than hiding AI implementation.

  • Technical teams need privacy-first frameworks that address our low trust levels (39% vs 79% global average), focusing on data protection, local processing options, and explainable AI systems.

  • Marketing and communications professionals must master honest AI messaging that manages customer expectations and responds to concerns authentically.

Rather than fighting our cautious nature, we should position it as creating higher-quality implementations, better customer relationships, and sustainable competitive advantages. Our cautious approach could also be smart positioning, as demanding AI disclosure and data protection puts pressure on companies to build better safeguards. This skepticism might actually help us avoid the worst AI pitfalls while still capturing productivity gains.

🐝 Save hours of work starting today.

Try Autohive free and build AI agents in minutes that tackle your most repetitive tasks - secure, private, and ready when you are.

🏛️ Government, Policy & Legal

AI music platforms scrape internet content without paying artists whilst NZ offers zero protection. Suno faces US and German lawsuits over copyright but operates freely here under "fair use" claims. NZ's unique Copyright Act automatically grants AI-generated song ownership to users, even when based on others' work.
5-min listen.

  • Our take: We've accidentally created a legal haven for AI music piracy where platforms can harvest Kiwi artists' work without compensation whilst competitors face court battles overseas. This regulatory gap could decimate our music industry's revenue streams and traditional artists’ motivations.

💼 Business & Industry

Four major Kiwi health organisations launch joint AI venture. Health Accelerator combines 500+ GP practices from Pegasus, Pinnacle, ProCare and Tū Ora Compass Health to deploy digital solutions nationwide. AI scribes like Heidi Health and NABLA Copilot automatically transcribe consultations, saving ~7 minutes per 15-minute consultation whilst automated admin processes handle inbox management and government funding claims.
9-min listen.

  • Our take: Smart move to pool resources across organisations creates national scale that individual private health organisations couldn't achieve alone. If this model works, expect rapid expansion as ROI becomes proven and other networks (not just in health) follow suit. Early wins in reducing administrative burden could transform primary care experiences across New Zealand.

Kiwi tourism breaks barriers with AI-powered trip planning. Tourism New Zealand's GuideGeek integration addresses booking hesitations like flight times and seasonal activities through personalised AI responses. Platform trained on extensive local data helps international visitors move from dreaming to actual trip planning.
3-min read.

  • Our take: AI democratises travel planning by removing knowledge barriers that previously deterred visitors. When tourism becomes more accessible through technology, destinations like New Zealand benefit from broader visitor diversity.

Sixty-year-old Datacom positions itself as AI transformation leader. From 1965 mainframes to AI agents writing code, NZ's original tech unicorn completed strategic restructuring into four business lines while investing in GPU capacity. The company maps AI opportunities across government ministries and builds industry-specific solutions for local councils.
4-min read.

  • Our take: Datacom's survival from punch cards to neural networks proves adaptability beats age in tech. Their focus on local solutions rather than global platforms could be the secret sauce for AI adoption in smaller markets.

🎓 Education & Society

Booking.com survey reveals 79% of Kiwis want AI trip planning despite trust concerns. Nearly half of local travellers already use AI for holiday planning whilst only 3% fully trust the technology. Air New Zealand's incoming CEO used ChatGPT to plan his entire three-week European family trip.
3-min read.

  • Our take: We're witnessing a fundamental shift in how Kiwis source travel information, from human personalities to algorithmic assistance. The cautious adoption suggests we value efficiency but haven't abandoned our healthy skepticism about digital advice. The travel industry's AI adoption is outpacing consumer confidence, a classic early-market dynamic. Smart operators will focus on transparency and human backup systems.

Auckland Uni builds AI pronunciation coach for te reo Māori learners. Computer Science researchers created an AI system that gives real-time feedback on Māori pronunciation, targeting one million users within five years. The platform received MBIE Smart Ideas funding and partnership with Te Hiku Media, whose CEO ranks among the world's top 100 AI innovators.
4-min read.

  • Our take: This flips the script on language revitalisation, instead of burdening Māori educators with endless pronunciation feedback, AI handles the repetitive work. More confident speakers means stronger cultural connections and reduced educator workload.

Auckland Uni scores $4m to build AI dementia prediction tool. Researchers aim for 85% accuracy in predicting dementia risk using AI for memory tests, blood results, genetics and MRI scans. The three-year MBIE-backed project includes partnerships with Siemens Healthcare and Singapore experts, targeting early detection before symptoms appear.
1-min read.

  • Our take: With 70,000 Kiwis living with dementia set to double by 2050, early prediction tech creates massive healthcare savings. If this hits target accuracy, it transforms from reactive treatment to proactive prevention planning.

🎙️ Latest from The AI Corner podcast

This week’s guest is Karrina Mountfort, founder of The AI Assembly and Webbased AI.

Karrina shares her journey from building enterprise systems to pioneering community‑focused AI in New Zealand. Discover why she reframes AI adoption as education and ethical licensing, and why real human connection matters just as much as the tech.

Listen on Spotify or YouTube 🎧

Subscribe on Spotify and YouTube to be notified of new episodes.

📚️ Levelling Up With AI

1️⃣ AI could spark NZ’s next SME boom. With middle layer jobs under threat, AI lets Kiwis start businesses faster and cheaper. Mass layoffs could be the spark for a surge in small business growth, rather than a collapse.
4-min read.

2️⃣ ChatGPT conversations are (now that’s a ‘were’) public if you share them. A Google trick is surfacing shared ChatGPT chats. Over 70,000 are already indexed. If you want privacy, do not share them.
1-min read.

3️⃣ Why AI’s unpredictability is an asset not a flaw. AI’s unpredictability unsettles some, but younger AI natives see it as a strength, which could benefit NZ’s business sector.
5-min read.

4️⃣ How simple AI apps are becoming big businesses. “Just wrapper” AI apps are growing into multimillion dollar companies. For NZ, they are a fast track to global products without building a model.
5-min read.

🌍 The News from Global

🔍 Fake AI news anchors go viral. From TikTok to X, synthetic presenters deliver false stories with professional polish, raising concerns about disinformation while helping journalists in oppressive regimes stay anonymous.
3-min read.

🌏 China proposes global AI governance framework. Premier Li Qiang calls for international coordination as China's AI industry hits $84B value with 5,000+ companies, while the US pursues its own dominance strategy.
6-min read.

🎬 Hollywood embraces AI filmmaking tools. Studios now use AI for voice cloning and visual effects, with Netflix completing sequences 10x faster and at a fraction of traditional costs after recent union strikes secured AI guardrails.
4-min read.

Tech Updates You Should Know

  • OpenAI: Launched 'Study Mode' in ChatGPT to guide users through learning with interactive steps and adaptive quizzes, designed to reduce cheating; announced Stargate Norway, a renewable-energy-powered European data centre; CEO Sam Altman warned about weak privacy protections and AI-driven voice scams, called for minimal regulation; revealed GPT-5 mini and nano versions coming to API and edge devices, with GPT-5 Prime features like 'Video Chat' and 'Mobius' spotted in leaks; facing a court battle with The New York Times over storing user chats; OpenAI’s model narrowly lost to human coder in AtCoder Finals by 9.5%.

  • Google: Launched AI-powered features in Chrome and Search including store summaries, 'Search Live' for real-time Q&A, and 'Canvas' study planner; updated NotebookLM with 'Video Overviews' for transforming notes into AI-generated narrated slides; debuted Opal, a no-code tool for building mini apps; expanded Google Photos with AI photo-to-video and remix tools; Google DeepMind launched AlphaEarth Foundations to generate environmental maps from Earth observation data.

  • Meta: Appointed former OpenAI researcher Shengjia Zhao as Chief Scientist of its new Superintelligence Labs; experimenting with AI in hiring processes; remains non-signatory to EU’s AI Code of Practice citing legal ambiguities.

  • Microsoft: Introduced 'Copilot Mode' in Edge browser with AI that searches tabs, summarises, and assists with tasks using history, potentially moving to subscription model; tested 'Smart Mode' for Notepad and Microsoft 365 Copilot using GPT-5 to auto-select best models for user tasks; nearing long-term deal with OpenAI to secure tech access beyond AGI milestone; Copilot Vision for screen analysis continues Windows 11 AI integration.

  • Anthropic: Capped Claude Code usage for some users due to overwhelming demand; CEO Dario Amodei condemned Meta's aggressive talent poaching; nearing close of $5B funding round led by Iconiq Capital, boosting valuation to $170B from $61.5B; part of coalition concerned about chain-of-thought AI monitoring reliability.

  • xAI: Opened waitlist for 'Imagine', a multimodal feature enabling image and video generation; planning infrastructure on scale of 50M NVIDIA GPUs by 2025.

  • Amazon: Finalised multiyear licensing deal with The New York Times worth up to $25M annually to train AI models; involved in multiple AI collaborations, including joint Llama AI work with Meta.

  • NVIDIA: Chips banned in U.S. are resurfacing in China through black market channels, with ~$1B worth reportedly smuggled.

📅 AI Events in New Zealand

A humble 15 AI events happening this week across the country, but don’t mistake low quantity for low quality.

This week’s featured event:

📅  Promote your event with us. Reply to let us know.

💼 AI Roles Around Aotearoa

Picklist of 🌶️ HOT 🌶️ new roles in AI this week.

💼  Promote your job with us. Reply to let us know.

🤦 ️ AI Fail Of The Week

We all love AI, but it’s certainly far from perfect 🤔 …

Asking AI to produce a ‘Where’s Waldo?’ scene.

👋 Mike & Erin