AI in Creative Industries in Aotearoa New Zealand: A Living Whitepaper

Introduction

AI adoption across Aotearoa New Zealand’s creative industries is broadening from individual experimentation into organisational practice. The latest evidence does not suggest a sudden shift to fully automated creative production. Instead, it points to a more incremental model: AI is being used to accelerate research, translation, transcription, ideation, optimisation, accessibility, administration and audience discovery, while human creators remain responsible for judgement, cultural context and final outputs.

The latest national baseline remains Manatū Taonga’s 2025 Cultural Participation Survey. It found that 69% of creators use digital tools, representing 28% of New Zealand adults, and that 65% of those digital-tool users use generative AI. Nearly half use it to explore or improve ideas, 34% use it to generate or produce creative work, and 14% use it to share work more widely or improve accessibility. (mch.govt.nz)

Since the July 13 update, the most important development has been the further institutionalisation of AI. Funding agencies are embedding disclosure and cultural-impact questions into applications; publishers are moving from pilots to AI-enabled operating programmes; marketing organisations are treating AI-driven discovery as a strategic issue; and public concern is expanding from copyright and privacy to employment, sustainability and human accountability.

The overall picture is therefore one of broad but uneven adoption, increasingly formal governance, and unresolved questions about rights, value distribution and the future of entry-level creative work.

Executive Snapshot

• Adoption is already mainstream among digitally active creators. Generative AI is being used most often for ideation, refinement and administrative support rather than replacing entire creative processes. (mch.govt.nz)
• Workflow AI remains more mature than synthetic public-facing content. The strongest New Zealand case studies involve translation, document analysis, SEO, transcription, summarisation, audience analytics and content transformation.
• Public funding is becoming a governance mechanism. NZFC, NZ On Air and Creative New Zealand now require or encourage disclosure of AI use and expect applicants to address cultural authenticity, creative integrity, IP, transparency and risk. (nzfilm.co.nz)
• Journalism is the most documented adoption sector. Google-supported pilots have reported major reductions in translation time, increased story output and improved newsroom SEO capability. (blog.google)
• Marketing is moving from content generation to AI-shaped discoverability. Trade Me is a prominent local example of AI being embedded across planning, creative development, optimisation, modelling and reporting. (marketing.org.nz)
• Trust is becoming a practical adoption constraint. One NZ reports that 62% of New Zealanders would stop using a product or service if they were concerned about how AI was being used, while 68% would be more comfortable if a human option remained available. (media.one.nz)
• Sustainability has entered the AI trust debate. One NZ’s July research found that 45% of New Zealanders are concerned about the environmental impact of business AI use, rising to 63% among 18–24-year-olds. (media.one.nz)
• Entry-level creative work is an emerging pressure point. Massey University research indicates that AI may be replacing some low-paid, freelance and unpaid work traditionally used to build early-career portfolios and networks. (massey.ac.nz)
• The games sector remains a major creative-technology frontier. The latest GDSR data records $829 million in combined recipient-studio revenue, 194 games in development and 98% export revenue, although public data does not yet isolate AI adoption. (nzonair.govt.nz)

What Changed Since the July 13, 2026 Update

1. AI capability programmes in publishing moved from pilots toward scale

On July 28, Google News Partnerships published results from AI workshops involving New Zealand and Pacific publishers. The reported pilots included:

• Pacific Media Network: A Gemini-based translation tool reduced the reported time needed to translate daily news bulletins into Tongan and Cook Islands Māori from up to three hours to approximately 15 minutes.
• The Central App: A tool designed to identify story angles in council meeting transcripts reportedly enabled two reporters to double their story output.
• Newsroom NZ: An SEO tool improved reporters’ confidence in producing search metadata, with self-rated confidence increasing from 2.7 to 4.3 out of five.
• An 18-month AI programme: Google announced a broader programme for New Zealand and Pacific publishers focused on production, distribution and monetisation. (blog.google)

This represents a shift from isolated newsroom experimentation toward structured capability-building and repeatable AI workflows.

2. Sustainability became part of the AI trust conversation

One NZ’s AI Trust Report 2026: Sustainability Edition, released July 29, found that public concerns are no longer limited to data privacy or automation. Respondents identified electricity use, water consumption, electronic waste and carbon emissions as significant concerns associated with AI infrastructure.

The same research found that:

• 65% were concerned about AI replacing human jobs.
• 62% were concerned about reduced human interaction in customer service.
• 60% were concerned about over-reliance on AI at the expense of human oversight.
• 59% were concerned about organisations using personal information without consent. (media.one.nz)

For creative organisations, this widens the definition of responsible AI. Governance now includes not only whether an output is accurate or lawful, but also whether the technology is socially legitimate, environmentally defensible and supportive of human work.

3. Music funding processes now explicitly address AI-generated songs

The current NZ On Air music funding information requires applicants to consider how AI and generative AI will be used in song creation. It links applicants to formal AI Content Creator Guidance and states that AI use will be assessed as part of the funding process. (nzonair.govt.nz)

The guidance assesses proposed AI use against:

• Cultural authenticity, including te ao Māori and mātauranga Māori.
• Creative integrity and artistic vision.
• Content distinctiveness and market gaps.
• Risk management.
• Diversity, competition and accessibility.
• Value and cost efficiency.
• Capability and technical expertise.
• Transparency.
• Ethical and wider industry impacts. (nzonair.govt.nz)

This is a notable development because music is now being treated not simply as a sector experimenting with new tools, but as a publicly supported creative field requiring explicit AI disclosure and assessment.

4. Marketing’s AI-first operating model became more visible

The Marketing Association’s August 3 report on its July 28 Digital Day Out reinforced the view that the next competitive advantage will not come from access to AI tools alone. Instead, organisations will need the culture, training and operating structures to use AI consistently.

Trade Me was presented as a leading New Zealand example. Its Marketing Guild reportedly embedded AI across:

• Strategy and briefing.
• Creative development.
• Data modelling.
• Copy optimisation.
• Performance reporting.
• Anomaly detection.

The company’s stated operating principle was “co-pilot, not autopilot”: AI should augment human capability rather than replace human judgement. (marketing.org.nz)

IAB New Zealand’s August 13 Discovery: AI & Search Summit further signalled that AI is changing how brands, publishers and creative work are discovered. The event brought together representatives from OpenAI, Google, Kantar, Taboola and New Zealand search and marketing firms to examine AI-generated answers, agent-mediated discovery and changing consumer behaviour. (iab.org.nz)

5. The copyright timetable became clearer

MBIE’s copyright update, last revised on August 7, confirms that the Government has asked the Minister of Commerce and Consumer Affairs to report by March 31, 2027 on a possible copyright framework for generative AI in New Zealand. The current copyright reforms do not themselves resolve training-data, licensing, AI-output or creator-consent questions. (mbie.govt.nz)

The practical implication is that New Zealand’s creative industries remain in a period of legal uncertainty. Organisations must make AI decisions under existing copyright, contract, privacy, consumer and cultural-protection frameworks while awaiting more specific policy treatment.

Current State by Subsector

1. Journalism and Publishing

Journalism continues to provide the clearest public evidence of operational AI adoption in New Zealand.

The AUT Journalism, Media and Democracy research centre’s 2026 baseline report found that AI tools are commonly used in newsrooms for research, transcription, summarisation, content transformation, audience analytics and other production tasks. The report also found uneven disclosure about when and how AI is used. (openrepository.aut.ac.nz)

Publicly documented use cases

• Stuff’s Democracy.AI has been used to scan public documents such as council minutes and identify potential local stories.
• NZME’s Bidi assists BusinessDesk with rewriting and publishing NZX announcement stories.
• NZ Herald’s Polaris supports homepage curation and recommendation.
• BusinessDesk’s Today in Business used AI to draft and voice a podcast from journalist-created material, with editorial checking before publication.
• Google-supported pilots have added translation, council-document analysis and SEO assistance to the sector. (openrepository.aut.ac.nz)

Human accountability remains central

RNZ’s 2026 principles state that the organisation generally will not knowingly publish or broadcast material created by generative AI. It permits assistive uses such as research, brainstorming, administration, summarisation and transformation of already-created content, provided humans remain accountable. (rnz.co.nz)

TVNZ’s principles similarly permit AI for research, transcription, translation, audio enhancement and selected post-production tasks, while requiring manager approval, journalist verification and senior editorial review. TVNZ says audiences will be informed when AI plays a meaningful role in editorial news. (corporate.tvnz.co.nz)

Assessment

New Zealand journalism is moving toward AI-assisted scale without fully automated editorial authorship. The strongest use cases increase coverage or reduce routine workload. However, public trust, provenance and transparency remain limiting factors. AUT’s 2026 Trust in News research found that 60% of New Zealanders remain uncomfortable with news mostly produced by AI. (aut.ac.nz)

2. Screen, Film, Television and VFX

Screen remains the sector with the most developed formal AI governance.

NZFC funding guidance now requires applicants to explain proposed AI use and address:

• Respect for te ao Māori and mātauranga Māori.
• Cultural authenticity and appropriate consultation.
• Preservation of human creative leadership.
• Transparency to funders and audiences.
• Legal, ethical and wider industry effects. (nzfilm.co.nz)

NZFC’s wider AI principles emphasise human talent, culture, accountability, data protection, intellectual property and responsible oversight. (nzfilm.co.nz)

TVNZ has also stated that local producers should notify the broadcaster about planned generative AI use, particularly where it may affect on-screen talent, key visuals or other audience-facing material. (corporate.tvnz.co.nz)

Capability and experimentation

The Aotearoa AI and Creativity Summit, held in Auckland and Wellington in May, brought together creators, educators, technology organisations and sector leaders to discuss synthetic storytelling, copyright, attribution, social licence and creative practice. Its associated One Minute AI Film Festival provided a visible experimentation layer for AI-assisted filmmaking. (aicreativeindustries.nz)

The proposed Aotearoa Creative Artificial Intelligence Research Institute, led by Wētā FX, remains an important infrastructure signal. The concept includes research into computer vision, generative models, digital twins, physically plausible datasets and AI rights management, with potential applications extending beyond entertainment into physical simulation, medical imaging and manufacturing. The latest public MBIE documentation describes it as one of the concepts selected for further development under the national AI Research Platform. (mbie.govt.nz)

Assessment

Screen-sector adoption is likely to remain concentrated in pre-visualisation, script analysis, asset development, visual effects, post-production and production planning. The strongest governance trend is the movement from voluntary principles to funding-stage disclosure and assessment.

3. Advertising, Marketing and Commercial Creative

Commercial creative is likely to be the fastest-normalising segment of the market.

The Marketing Association’s Trade Me case study shows AI being embedded across the complete marketing lifecycle rather than confined to copy generation. IAB New Zealand’s programme of events likewise frames AI as a change to creativity, search, audience discovery, measurement and brand strategy. (marketing.org.nz)

The market is increasingly separating into two broad layers:

• High-volume, lower-stakes work: social content, performance creative, variant generation, copy adaptation and routine production.
• High-craft, higher-stakes work: brand platforms, broadcast campaigns, distinctive visual identity, cultural storytelling and work requiring strong human judgement.

AI is also changing the commercial meaning of “discoverability”. Brands are beginning to optimise not only for traditional search engines and social platforms, but also for AI-generated answers, recommendation systems and agent-mediated purchasing journeys.

Rights and likeness

The 2026 Huffer controversy demonstrated how quickly commercial AI use can become a rights and trust issue. Models alleged that AI-generated campaign images resembled them and other models without sufficient notice, consent or compensation. Huffer denied wrongdoing and described its practices as involving computer-assisted technologies. Legal commentary cited by 1News highlighted the importance of contracts, consumer law, copyright and the lack of a standalone New Zealand personality or likeness right. (1news.co.nz)

Assessment

The commercial sector is moving fastest toward AI-native operating models, but the most defensible organisations will be those that combine production efficiency with clear consent, transparent contracts, brand accountability and human creative direction.

4. Music

Music remains the creative subsector where AI adoption, economic risk and rights concerns are most tightly connected.

APRA AMCOS research continues to provide the main Australasian benchmark. It found that 54% of surveyed creators believed AI could assist human creativity, while 82% worried that AI could threaten their ability to earn a living. The research estimated that 23% of music creators’ revenues could be at risk by 2028 without effective licensing arrangements. (mch.govt.nz)

The most recent New Zealand development is procedural: NZ On Air’s current music funding process explicitly asks how AI or generative AI will be used in song creation and assesses the proposal against cultural, creative, legal and industry criteria. (nzonair.govt.nz)

Main adoption areas

• Songwriting ideation and experimentation.
• Sound design and arrangement.
• Translation and accessibility.
• Marketing assets and promotional content.
• Administrative and funding applications.
• Early-stage demos and visualisation.

Main risks

• Unlicensed training data.
• Unclear ownership of AI-assisted compositions.
• Voice and likeness imitation.
• Reduced demand for session musicians, producers and early-career contributors.
• Cultural misrepresentation or misuse of Māori and Pacific musical knowledge.
• Difficulty distinguishing human-led work from predominantly synthetic output.

Assessment

Music is likely to remain a cautious-adoption sector. Creators may use AI privately or experimentally while resisting public release where provenance, consent, compensation or cultural integrity cannot be demonstrated.

5. Games and Creative Technology

Games and interactive media remain New Zealand’s strongest creative-technology growth frontier.

NZ On Air’s 2026 GDSR data records:

• $829 million in combined recipient-studio revenue in 2025/26.
• 17% year-on-year revenue growth.
• 43 recipient studios, up from 40.
• 194 games in development, up from 170.
• 98% of recipient revenue generated from exports.
• 90% of recipient studios classified as SMEs. (nzonair.govt.nz)

The data does not separately measure AI adoption. However, the sector’s export orientation, technical intensity and reliance on small teams create strong incentives to adopt AI for asset pipelines, prototyping, testing, localisation, simulation, animation and production management.

The Creative Tech Accelerator, co-designed by Microsoft, NZIST and Seen Ventures, provides an example of capability-building. Its 12-week programme combines emerging creative talent, industry briefs and AI-enabled production techniques, with an emphasis on using AI as an integrated creative partner rather than a standalone tool. (nzist.ac.nz)

Assessment

Games and VFX may generate the largest long-term productivity gains from AI, but they also face significant questions about artistic labour, data provenance, training pipelines and the preservation of distinctive human-made worlds.

6. Arts Funding, Education and Creative Institutions

Creative New Zealand’s guidance makes clear that AI use will not, by itself, disadvantage an applicant. It asks creators to disclose AI use, lead with their own voice, review outputs, respect IP, protect private information and seek appropriate cultural advice for projects involving Indigenous, Pacific or other communities. (creativenz.govt.nz)

This approach is broadly consistent with NZFC and NZ On Air: adoption is permitted, but responsibility remains with the creator or applicant.

Workforce and entry-level roles

Massey University’s pilot study, based on interviews with ten creative practitioners, found that AI was most commonly used for low-value administration, followed by ideation and experimentation. Participants generally regarded AI as a time saver rather than a cost saver, and most would not deploy AI outputs commercially without extensive human review. (massey.ac.nz)

The study identified a more subtle labour risk: AI may replace low-paid, freelance or unpaid work that previously allowed emerging creatives to build experience, professional relationships and portfolios. The researchers described this as the possible loss of “gateway” work rather than immediate large-scale job destruction. (massey.ac.nz)

This is particularly important for New Zealand’s small creative sectors, where volunteer labour, portfolio-building and informal networks often support entry into professional practice.

Research Evidence and Limitations

The evidence base is improving, but it remains uneven.

Strongest available evidence

• National creator adoption: Manatū Taonga’s Cultural Participation Survey.
• Newsroom practice: AUT JMAD’s 2026 baseline report.
• Creative-sector practitioner experience: Massey University’s ten-interview pilot study.
• Public funding governance: NZFC, NZ On Air and Creative New Zealand guidance.
• Audience trust: AUT Trust in News and One NZ AI Trust research.
• Games-sector scale: NZ On Air’s GDSR data.

Important limitations

• There is not yet a comprehensive, sector-by-sector measurement of AI adoption across New Zealand’s creative industries.
• Many public case studies are supplied by technology vendors, funders or participating organisations and may emphasise positive outcomes.
• Public reporting is stronger for journalism and marketing than for independent arts, publishing, music production and small screen businesses.
• Productivity gains are not consistently translating into improved profitability.
• AI adoption is often embedded inside software platforms and may not be recognised or disclosed as “AI”.
• There is still little public evidence about how AI is affecting Māori-led creative practice, Pacific arts or community-controlled cultural data.

Cross-Sector Trends

1. Adoption is becoming institutional rather than experimental

AI is now appearing in funding applications, broadcaster principles, newsroom programmes, marketing operating models and creative-skills training. The key question is shifting from whether AI may be used to how it should be governed.

2. Assistive AI remains the dominant model

New Zealand’s public case studies continue to favour research, translation, transcription, summarisation, optimisation, accessibility and ideation. Human-led production remains the preferred model in journalism, screen funding and public media.

3. The biggest labour risk is at the entry level

The immediate concern is not necessarily mass replacement of established practitioners. It is the disappearance of low-paid and unpaid opportunities through which early-career creatives traditionally acquire skills, networks and professional credibility.

4. Cultural integrity is becoming a core AI capability

For Aotearoa, responsible AI cannot be reduced to accuracy, efficiency or copyright compliance. Public agencies increasingly expect cultural consultation, respect for mātauranga Māori, attention to te reo Māori and meaningful consideration of Māori data sovereignty.

5. Discoverability is now part of creative production

The value of creative work increasingly depends on whether it can be found and recommended by search engines, social platforms and AI systems. Marketing, publishing and media organisations are therefore adapting not only how they make content, but how they structure and distribute it.

6. Trust is moving from principle to operating requirement

Human review, clear disclosure, consent, provenance, accountability and access to human support are becoming practical conditions for public acceptance. Sustainability is now joining those requirements.

Outlook

Over the next 12 months, the most important developments are likely to be:

• Further integration of AI disclosure into funding and commissioning processes.
• Expansion of Google’s AI programme for New Zealand and Pacific publishers.
• Greater use of AI for translation, accessibility and multilingual content.
• More commercial experimentation with AI-generated advertising and synthetic talent.
• Continued pressure for licensing and consent mechanisms in music and publishing.
• More formal debate over likeness, voice, digital replicas and cultural IP.
• Further development of the proposed national AI research platform and its creative-AI component.
• Greater scrutiny of the effect of AI on early-career employment and creative education.

Conclusion

As of August 18, 2026, AI adoption in Aotearoa New Zealand’s creative industries is best described as broad, workflow-led and increasingly governed.

The July-to-August period did not produce a fundamental change in the adoption baseline. It did, however, strengthen the evidence that AI is becoming part of normal organisational infrastructure. Publishers are building repeatable AI services. Marketing teams are redesigning operating models. Music funders are asking applicants to disclose AI use. Screen agencies are incorporating AI into project assessment. Public trust research is expanding to cover employment, human interaction and environmental impact.

The sector is not moving uniformly toward autonomous content generation. Instead, it is developing a distinctly New Zealand pattern of human-led, culturally aware and commercially pragmatic adoption.

The central strategic challenge is now distributional: who benefits from AI-enabled productivity, who bears the cost, and whether the gains strengthen or weaken the creative ecosystem. New Zealand’s most successful adopters will not necessarily be those using the greatest volume of AI. They will be those able to demonstrate that AI improves creative capability while protecting consent, cultural integrity, human accountability, professional development and audience trust.