How does incident reporting work under the #AIAct? It's a bit more complex than you might imagine. The precise rules vary depending on whether the AI system is "high risk" or not, whether the incident itself is "serious", whether you are a provider or deployer, and, if a provider, whether you provide the impacted AI system or a #GPAI model integrated into it. Got all that? Don't worry if not - the Digiphile infographic below should help:
I wish they had a form for both developers and consumer that is more user friendly for reporting said incidents https://artificialintelligenceact.eu/article/73/
We offer AI incidence response training and preparation from a business impact perspective.
Less reporting burden please.
Thank you Phil Lee for this infographic. It breaks down the law related to AI incident reporting in a manner that is easy to understand.
Angelo Ganno (Sagala) Anne-Claire DUBOIS
#sobs
Thanks - that’s very useful!
Founder of the Test Community Network, Speaker, Podcaster, Teacher, Fractional Chief Marketing Officer, and Edtech Product Marketing & Events Consultant
1moPhil, I really appreciate all the excellent content you're sharing on the topic of the AI Act. Do you know of any latest documents that summarises the current UK approach? My reading seems to suggest that we're not heading in quite the same direction, but if you know differently, I'd appreciate you pointing me in the right direction. Thanks, Tim