[SystemSafety] AI and safety

[Olwen Morgan]

Her database is an excellent contribution to the debate on AI and safety.

Only when we build up large lists of cock-ups like that, will we be able 
to spot situational patterns where the reach of AI specialists has 
exceeded their grasp (which, AFAI can see, has happened depressingly 
often in the repeated cycles of hype that AI research undergoes).

Olwen


On 09/11/2018 20:04, Robert P. Schaefer wrote:
> Hi,
>
>  just pointing out the websight of an AI researcher who is involved in 
> AI safety:
>
> https://vkrakovna.wordpress.com
>
> https://vkrakovna.wordpress.com/2018/11/01/discussion-on-the-machine-learning-approach-to-ai-safety/
>
> And has a database of learning AI’s that when given a specification, 
> make undesireable choices:
>
> https://docs.google.com/spreadsheets/u/1/d/e/2PACX-1vRPiprOaC3HsCf5Tuum8bRfzYUiKLRqJmbOoC-32JorNdfyTiRRsR7Ea5eWtvsWzuxo8bjOxCG84dAg/pubhtml 
>
>
> bob s.
> research engineer
> mit
>
> _______________________________________________
> The System Safety Mailing List
> systemsafety at TechFak.Uni-Bielefeld.DE
> Manage your subscription: https://lists.techfak.uni-bielefeld.de/mailman/listinfo/systemsafety
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20181109/c01401ae/attachment-0001.html>