- The EU AI Act has just been passed – the world’s first comprehensive law to regulate AI. Taking a risk based approach, it will have extraterritorial effect just like GDPR, and is likely to have a similar influence on legislators around the world. Check out IAPP’s cheat sheet plus this AI Act 101 resource.
- On a related note, we commented on LinkedIn about clarification from the Technology Minister and Attorney-General, Judith Collins, that “there will be no extra regulation at this stage” of artificial intelligence in NZ. It appears the Government is currently developing some form of “AI framework”, though details remain scarce. We shared our wish list of the improvements we’d like to see made to our privacy and human rights regimes if NZ is not going down the AI-specific regulation path.
- Our LinkedIn post about Foodstuff’s facial recognition trial also prompted lots of discussion about the pros and cons of facial recognition. Having reviewed the Privacy Impact Assessment, we noted the absence of important information like the model and training data used. This demonstrates sone of the limitations of using PIAs to assess AI risks and we shared a link to the Algorithmic Impact Assessment toolkit that Frith developed for government agencies.
- This week completed the first of our live virtual AI Governance Professional training sessions as an Official Training Partner of the IAPP, having sold out all our spaces. Registrations for the next training session in June are now open and filling up fast, so register now to avoid missing out.