[SystemSafety] AI and safety

[Peter Bernard Ladkin]

On 2018-11-09 21:04 , Robert P. Schaefer wrote:
> Hi,
> 
>  just pointing out the websight of an AI researcher who is involved in AI safety:

The term "AI safety" may or may not have anything to do with safety as we conceive it.

For example,

[begin quote]
Short-term: This work involves immediately practical safety risks in deploying machine learning
systems. These include data poisoning, training set inference, lack of model interpretability, and
undesirable model bias.
Mid-term: This work targets potential safety risks of future AI systems that are more powerful and
more broadly deployed than those used today. Relevant problems in this space include *scalably
specifying and supervising reward-based learning, preventing unwanted side effects, safely
generalizing out of domain, and ensuring that systems remain under our control.
Long-term: This theoretical work addresses the risks posed by artificially engineered
(super)intelligences. It asks, for example, how we might ensure that a system is aligned with our
values, and proposes procedures for conserving this alignment while supporting recursive
self-improvement.

[end quote]

There is no connection made from these features which consitute "AI safety" to harm caused to or the
environment, and damage to things, avoidance of which is the usual definition of safety.

I wonder, are they really talking about reliability? You know, the usual confusion we have been
pointing out for three decades. All of the phenomena above obviously affect system reliability.

PBL

Prof. Peter Bernard Ladkin, Bielefeld, Germany
MoreInCommon
Je suis Charlie
Tel+msg +49 (0)521 880 7319  www.rvs-bi.de





-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20181110/a38ccd8e/attachment.sig>