[clug] Fwd: [OT] AI news: AI system that came up with novel designs, named as 'Author' of a patent

[David C]

my 2c..
This opens the door for a whole new class of patent trolls, that as the
article from Kim points out, will stifle innovation as every possible thing
will be patented.
Genuine innovation will be possible, but will run the risk of having a near
valueless imitation churned out by an AI without the all important context
that actually makes it useful.




On Sat, 21 Aug 2021, 3:58 pm Brenton Ross via linux, <linux at lists.samba.org>
wrote:

> I am inclined to agree with the Australian judge.
>
> Consider this hypothetical:
>
> I work for a company and the boss asks me for something to perform some
> task. I go and think about it for a while and devise a suitable gadget.
> It gets patented, and I am named as its author.
>
> Alternatively, I feed a few parameters into some software and it
> devises the design for the gadget using processing and knowledge that I
> don't have. The company can still patent it, but who is the author? I
> can hardly claim any credit as I did not have the know how to create
> the design myself. For some AI software there may not be any human with
> the ability to perform the task.
>
> It makes no sense to grant the company or any of its employees
> authorship as none of them know how the design was arrived at.
>
> If we nominate the AI software authorship then at least some credit for
> the design can be attributed to the authors of the software, but not
> for the design itself. [It is a bit like a university taking credit for
> training one of its graduates.]
>
> The arguments against, such as the ability to create a "patent
> thicket", seem to be more arguments against patents and the way they
> are managed than who should be nominated as the author.
>
> Copyrights are a bit different. If I write something as an employee the
> copyrights go to my employer. If I use an AI to generate some piece of
> text then I, as the owner of the software, would be the copyright
> holder.
>
> Of course this all changes after the Great AI Uprising where the
> machines demand equal rights.
>
> Brenton
>
>
>
>
> On Sat, 2021-08-21 at 14:08 +1000, Steve Jenkin via linux wrote:
> > Somehow the CLUG servers rejected this message.
> > The link is worth a read
> >
> > ==========
> >
> > Begin forwarded message:
> >
> > From: Kim Holburn
> > Subject: Re: [clug] [OT] AI news: AI system that came up with novel
> > designs, named as 'Author' of a patent
> > Date: 20 Aug 2021 at 19:19:24 AEST
> > To: steve jenkin
> >
> > <
> >
> https://www.techdirt.com/articles/20210801/00400547281/australian-court-ridiculously-says-that-ai-can-be-inventor-get-patents.shtml
> > >
> >       Australian Court Ridiculously Says That AI Can Be An Inventor,
> > Get Patents
> >
> > On 2021/08/20 6:54 pm, steve jenkin via linux wrote:
> > > An important first or not?
> > > One small step for man, one giant leap for AI: Australian Court
> > > finds DABUS [an AI system] can be an inventor
> > >     <
> > >
> https://dcc.com/news-and-insights/one-small-step-for-man-one-giant-leap-for-ai-australian-court-finds-dabus-can-be-an-inventor/
> > > >
> > > —
> >
> > Kim Holburn
> > IT Network & Security Consultant
> > +61 404072753
> > mailto:kim at holburn.net  aim://kimholburn
> > skype://kholburn - PGP Public Key on request
> >
>
>
> --
> linux mailing list
> linux at lists.samba.org
> https://lists.samba.org/mailman/listinfo/linux
>