7
$\begingroup$

For example:

There is already the website for philosophical question, however here we can have more direct answers from the AI experts.

Should we allow such questions?

$\endgroup$
9
$\begingroup$

Yes.

If we send away everyone asking about philosophy; send everyone asking about feature selection for ANNs to data science and send everyone asking about AI research institutes to chat then there's really not so much left to talk about.

$\endgroup$
3
$\begingroup$

Yes, but I think many philosophical questions would be better off on Philosophy SE. It depends on the type of question. Questions that AI experts have mostly thought about (like "Why would an AI need to 'wipe out the human race'?") are better suited here, while questions that are tangentially related to AI but are really referring to "philosophical concepts" (robotic free will and AI creativity) are better left to the philosophy experts.

$\endgroup$
  • $\begingroup$ The problem is, philosophy as an armchair activity (i.e. without specific knowledge of the particular domain) is quite capable of being wrong by missing important subtleties. See this paper for some arguments of relevance to AI (scottaaronson.com/papers/philos.pdf). $\endgroup$ – NietzscheanAI Sep 7 '16 at 19:10
  • $\begingroup$ There's also the flip-side...AI experts who have no specific knowledge of the particular domain of philosophy are also quite capable of being wrong by missing important subtleties. It's important that people avoid "armchair" analysis by having a good background in both philosophy and AI development (which might be done by having overlap between the two SE sites). $\endgroup$ – Left SE On 10_6_19 Sep 11 '16 at 17:35
3
$\begingroup$
  1. The site proposal on the Area51:

    "For conceptual questions about life and challenges in a world where "cognitive" functions can be mimicked in purely digital environment."

    This very clearly includes the border to the phylosophy.

  2. Don't narrow the site topics.

    It results only a mass of people leaving the site disappointed after the closure of their first questions. With them, we lose not only their content, but also the content they could have made if their first experiences had been better.

    There is a so-named "common sense", what belongs to AI. It is what an ordinary people, who doesn't even know that a meta site exists, thinks what is AI. In my opinion, the topic of the site shouldn't ever be narrowed significantly below this "common sense".

  3. Pragmatical reasons.

    Currently we are absolutely not in the position where we could have the luxury to close questions. Later it may be better, but (1) and (2) will stay even then.


Note, I don't really like philosophical questions. I think the AI is more like on the engineering/science border as philosophical thing. If the site would seem to sink in the mess of endless philosophical debates, I would suggest to make a little limit (for example, to use the VtC as duplicate votes more rigorously), but this is not the case (now).

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .