Deepfake Law in New Zealand
New Zealand regulates deepfakes through the Harmful Digital Communications Act 2015, the Films, Videos, and Publications Classification Act, and Privacy Act 2020. AI-specific legislation is under consideration but not yet enacted.
- Status
- enacted
- Jurisdiction
- New Zealand
- Effective
- Jul 2015
- Statute
- Harmful Digital Communications Act 2015 + Privacy Act 2020
New Zealand's deepfake regulation operates through existing laws — primarily the Harmful Digital Communications Act 2015 and the Privacy Act 2020 — rather than dedicated AI legislation. Courts and regulators have interpreted these broadly to cover AI-generated content.
Key provisions
Harmful Digital Communications Act 2015. Creates a civil regime (administered by Netsafe) and criminal offenses for digital communications that cause serious emotional distress. Includes intimate visual recordings — explicitly interpreted to cover AI-generated imagery of identifiable persons. Civil remedies include takedown orders; criminal penalties up to two years.
Films, Videos, and Publications Classification Act 1993. Covers objectionable publications including AI-generated CSAM.
Privacy Act 2020. Biometric data provisions apply to voice and facial feature use in synthetic content depicting identified persons.
Electoral Act 1993. Covers false statements about candidates, applicable to election-context deepfakes.
Netsafe takedown regime
Netsafe is New Zealand's approved agency under HDCA. It operates a mediation-based takedown process:
- Complainant submits takedown request.
- Netsafe contacts the poster and/or platform.
- If unresolved, escalation to District Court for an order.
The process is fast by court-order standards (often days rather than months) and effective for cross-platform takedowns.
Practical implications
For organizations operating in New Zealand:
- Platforms: Netsafe cooperation is expected; voluntary compliance reduces formal court-order frequency.
- Creators of deepfake content: HDCA prosecutions for non-consensual intimate imagery have increased through 2023–2025.
- Enterprises: moderate regulatory burden; comparable to Australia but without eSafety's specialized deepfake-statute layer.