[ExI] AI definition

[Anders Sandberg]

On 2016-04-07 23:46, Keith Henson wrote:
> I would say we have AIs when they start getting author credit in 
> scientific papers.

What about this paper?
http://www.automaticstatistician.com/static/abcdoutput/02-solar.pdf
Note how it notices that the Maunder minimum 1643 to 1716 is odd.

http://www.automaticstatistician.com/index/

We are not there yet, but at least some AI now writes papers with some 
content.

-- 
Anders Sandberg
Future of Humanity Institute
Oxford Martin School
Oxford University