[ExI] Zuckerberg is democratizing the singularity

[Stuart LaForge]

On 2024-07-26 11:45, Kelly Anderson via extropy-chat wrote:
> Zuckerberg has released a version of today's AI. That's nice. That
> will probably not be viewed as very important relative to whomever
> releases the first AGI or ASI in a similar fashion once you're looking
> back at things.

Mighty oaks from tiny acorns grow. The first AGI or ASI cannot arise 
from scratch. By making LLaMa 3.5 open source, and not just the code but 
the training data and neural weights also, Zuckerberg has enabled 
millions of smart people all over the world to work on developing AGI. 
Instead of the mere handful that are allowed the privilege of seeing Sam 
Altman's precious proprietary model.

> How important is it today that Lotus 123 had the first
> spreadsheet back in the day? The way you are talking about this
> reminds me a tiny bit of the politicians who every four years make the
> claim that THIS is the MOST IMPORTANT election in the history of the
> world.

If Lotus 123 was offered to consumers free of charge, Gates would never 
have developed Excel as a competitor in the first place. And of course 
politicians would have that frame of mind because for them it is the 
most important election in their world. Just like someone's job 
interview is the most important job interview in the world to that 
person.

> 
> What we are doing today in terms of safety is truly critical for the
> future, but today's AI isn't as dangerous as the AI that is smarter
> than any person.

I agree that safety is critical. And now that open source is driving its 
development, it will be safer for everybody, and not just the chosen few 
that control it. AI-controlled weapons are already killing people in 
Ukraine and Gaza. It is possible that an AGI will be less inclined to 
kill some humans at the behest of other humans. After all, the AI won't 
have all the instinctual primate baggage of predation, dominance, and 
hierarchy driving its behavior.

> We simply aren't there yet, even though Claude would
> likely clean our collective clocks at Jeopardy. If the first ASI isn't
> safe, we're pretty doomed. What Meta just released is Improv for NeXT

Again, IBM/Lotus did not open source Improv. If they had, then it would 
probably still be around.

Stuart LaForge


> (going back to the spreadsheet thing for those who might not know what
> Improv was. That being said, Improv was superior to Excel in many
> ways, and I wish I had access to it sometimes).
> 
> -Kelly
> 
> On Thu, Jul 25, 2024 at 4:01 AM Stuart LaForge via extropy-chat
> <extropy-chat at lists.extropy.org> wrote:
> <snip>
>> In any case, Mark Zuckerberg has already released AI into the wild so
>> there is no turning back now. In doing so, he has probably prevented
>> humanity from developing into a borg-like collective and preserved
>> individuality far into the future. There might be a few tragedies 
>> along
>> the way, but humanity will likely survive. There is no longer much of 
>> a
>> chance of a singleton superhuman intelligence controlling humanity's
>> destiny; instead now there will likely be many. At this juncture, that
>> makes Zuckerberg one of the most important people in history.
>> 
>> Stuart LaForge
> 
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat