MIT may have done us a favor by putting together a compendium of risks associated with A.I. systems.
Named the "A.I. Risk Repository", there are presently 700 or so risks categorized in 23 frameworks by domain and cause, organized as a taxonomy for each of these characteristics.
The Causal taxonomy addresses 'how, when, and why' of risks.
The Domain taxonomy goes into 7 domains and 23 subdomains, so certainly some fine grain there.
YouTube, of course
This is a public resource, so naturally there's YouTube on what it's all about and how to use it.
There's a lot of stuff
If you go the link, given in the first paragraph, and scroll down a bit, you will be invited to wade into the database, working your way through the taxonomies. There's just a lot of stuff there, so give it a look.
Like this blog? You'll like my books also! Buy them at any online book retailer!
