[ExI] The Race to AGI is inevitable

[Brent Allsop]

Yay, another believer in Friendly AI
<https://canonizer.com/topic/16-Should-we-fear-AI/2-AI-can-only-be-friendly?is_tree_open=0>
.


On Fri, Jun 12, 2026 at 1:42 PM sjatkins via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> Humans seem quite efficient at persuading everyone to fight one another
> and to "get mine" assuming there isn't enough and never can be.  I don't
> see why malicious AGI would have much to do.
>
>
> But what if greater intelligence leads to more grounded and benevolent
> ethics?  For instance an intelligence seeing that only by maximizing other
> intelligent beings positive potential can its own choice and opportunity
> environment be as rich as possible and be thus enabled to maximize its own
> positive potential?
>
> What would a world be like where the goal is maximizing the positive
> potential of all humans and equivalently (or greater) intelligent beings?
>
>
> It would be very different from assumption of Scarcity, never enough for
> everyone, get yours before someone else does world.
>
> With AI much less AGI purportedly able to do so many jobs in a society
> that just assumes you must have such income and economic utility to be OK
> an increasing number of people greatly fear AI and technology.  They fear a
> world that has no place for them.
>
> A change of fundamental operating ethical basis and goal could greatly
> ameliorate this.    Removing IP laws that effectively create artificial
> scarcity and massive income inequality is another part.
>
>
> I believe and pretty much always have that we can reach a level of
> technological Abundance where all the needs and many of the desires of
> EVERYONE can be met with a small fraction of productive capacity.   But
> even with uber tech like machine phase nanotech we will never reach that
> point with our current assumptions and attitudes and the practices that
> grow out of them.
>
> Achieving the future of our dreams, achieving the fulfillment of humanity,
> is not just a matter of technology.  It requires a quite real and necessary
> change of consciousness.
>
>
>
>
> On Friday, June 12th, 2026 at 09:05, spike jones via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
> >
>
> >
>
> > -----Original Message-----
> > From: Ben Zaiboc <benzaiboc at proton.me>
> >
>
> > >
> > > _______________________________________________
> > >
> > >
> > >>... Ben the irony here is that our best defense against malicious AI
> comes
> > > from the defense establishment, for it understands and utilizes
> > > psyops, techniques for causing one's enemy to fight himself.  An
> > > emergent anti-human AI would convince humans to slay each other using
> > > the kind of psyops so highly effectively employed by Herr Hitler, who
> > > induced otherwise civilized Germans to slay millions of their own
> countrymen.
> > >
> > > An example of modern psyops would be to convince the masses that 1)
> > > they are poor, and 2) the reason they are poor is that billionaires
> > > are hogging up a disproportionate (and unfair) share of "the money"
> > > which is 3) a fixed quantity.  Result: the masses slay the richest
> > > person in town and redistribute "the wealth."  Result: they become
> > > poorer, for money is not a conserved quantity, and they just destroyed
> > > wealth, more than they liberated.  Result: they hunt the new richest
> > > guy in town.  Repeat until all are equally impoverished and begin to
> > > devour each other.  An anti-human AI could find ways to carry out
> psyops of that nature.
> > >
> > > The military uses psyops and understands the nature of such
> > > operations.  It has the best chances of employing countermeasures.
> > >
> > > spike
> >
>
> >
>
> > >...spike,
> >
>
> > >...You're basically saying that the human military has a good chance of
> outthinking AI.
> >
>
> > >...Think about that for a moment.
> >
>
> > ---
> > >...Ben
> >
>
> >
>
> > Ben what I meant was humans must understand how a malicious AI would
> destroy humanity.  It would convince humanity to fight itself using
> psyops.  It is not clear to me how to defend against that for most humans
> are underinformed on AI.  The countermeasures I referred to was not against
> AI but rather against humans who have been turned against humans.
> >
>
> > spike
> >
>
> >
>
> > _______________________________________________
> > extropy-chat mailing list
> > extropy-chat at lists.extropy.org
> > http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
> >
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20260612/8cc86316/attachment.htm>