2015-10-30 19:20 GMT+01:00 Brent Meeker <[email protected]>: > > > On 10/30/2015 9:30 AM, Quentin Anciaux wrote: > > > > 2015-10-30 17:13 GMT+01:00 Quentin Anciaux <[email protected]>: > >> >> >> 2015-10-30 17:01 GMT+01:00 John Clark < <[email protected]> >> [email protected]>: >> >>> On Thu, Oct 29, 2015 at 11:55 AM, Quentin Anciaux < <[email protected]> >>> [email protected]> wrote: >>> >>> >> >>>>> And I repeat, If the microprocessor made of matter that obeys the >>>>> laws of physics can't sense **any** information in the AI program then the >>>>> AI program is not running, it's not intelligent it's just a inert list of >>>>> instructions DOING nothing. >>>>> >>>> >>>> > >>>> Re-read what is written above... >>>> >>> >>> OK >>> >>> : >>> >>> " >>> And I repeat, If the microprocessor made of matter that obeys the laws >>> of physics can't sense **any** information in the AI program then the AI >>> program is not running, it's not intelligent it's just a inert list of >>> instructions DOING nothing. >>> " >>> >>> >>> >>>> > >>>>>> how can that affect if ? how can it knows that external world ?* >>>>>> >>>>> >>>>> >>>>> >> >>>>> It could have memories of that external world before the sensors were >>>>> detached, if they were never attached then it could have no knowledge of >>>>> that world >>>>> >>>> >>>> > >>>> That's the point... as no information of that "external" world is fed >>>> to it. >>>> >>> >>> Not true. The AI's physical memory banks are in that external world and >>> the information in it is sure as hell fed into it, as are the results of >>> calculations made by the physical microprocessors that are also in that >>> external physical world. >>> >> >> That's *not* information *about the physical world*... you are confusing >> level as usual. >> > > I'll try again... fooling myself again in believing you're honest here... > with a simple real life example. > > So let's pretend our "AI" is in fact a Nintendo Entertainment System > game... that game, can be run on a physical NES, or can be run in an > emulator running on a physical computer... or on an emulator running on an > emulator running on a physical computer... > > From the POV of the game, it is unable to distinguish those cases because > all informations it has from the substrate(machine running it) is the same, > the NES game cannot know it is not really running on a physical NES... > > Same thing with the AI, the information it has access to, it cannot tell > the ontological status of what that information represent... those > informations could come from a really real ontological physical world... or > an emulation of the really real ontological physical world or an emulation > of an emulation of the really real ontological physical world.... it has > no way to decide the ontological status of that "external world". > > > I think you are confusing a virtual world with an AI. >
No I'm not. > An AI must be embedded in a world, a context, which it interacts with. > where this "must" come from ? an AI has to have a context sure... a world "as in our everyday world"... why would it ? > It can only be intelligent in the sense of interacting intelligently. > Either it is conscious or it is not... if it is conscious, it knows it, it doesn't care if you see it behaving "intelligently" or not. > A virtual world can't be intelligent. > What are you talking about ? I'm talking about the ability for an AI to give an ontological status about the information it has on something external to it... it has simply no way of doing it, as my example with the game program illustrate it, the program cannot know if it is run directly on the "metal" or on an emulation of it... which means from it's POV, the direct "metal" or a virtual "metal" has the same "realness"... which makes the "metal" from it's POV an hypothetical (it can or not be ontological, the AI as simply no way to tell). Quentin > An AI will have some internal model of the world within which it interacts > and that model has an ontology. > > Brent > > -- > You received this message because you are subscribed to the Google Groups > "Everything List" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/everything-list. > For more options, visit https://groups.google.com/d/optout. > -- All those moments will be lost in time, like tears in rain. (Roy Batty/Rutger Hauer) -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
