(1/n)
A well written article.
However, regarding the conclusion, you should have stuck with your gut feeling, IMHO:
"How #ChatGPT is one step closer to becoming an Uber-AI."
I must admit, when I first heard the news that ChatGPT, #OpenAI’s #chatbot, was finally getting connected to the web, my imagination ran wild. Images of #dystopian future scenarios flooded my mind, reminiscent of the fearsome #Skynet from the #Terminator movies..."
It might neither be a #Skynet, nor...
(2/n)
"... #PKD's #SecondVariety, and not even #TheMatrix, but just as #surveillance technology has surpassed what was possible in #Orwell's #1984, in particular in the #PRC, we are moving beyond #Fahrenheit451, too.
👉The biggest trouble is that the ancient #Viking #Norns seem to be weaving all those #dystopian fates of humanity into one.👈
In the May 15th #TheEconomist's Editor's Picks podcast, they were looking historically at the impact on productivity growth...
(3/3)
...and the adoption rates for shiny new techs like trains, electricity, and autmatic telephone switchboards.
To cut a long (but interesting) story short: the adoption of new technologies took way longer than anticipated, and worker displacement did take place but was more than offset.
All very solid arguments. except:
1) never before has a technology gained almost global traction simultaneously (poor people w/o internet access an #NorthKorea being left out) and...
(4/n)
...b) has any technology or being ever been able to access almost all human knowledge simultaneously,
and
c)...has it been able to process all that knowledge without any significant time delay and...
d)...has it has a negligible cost for a huge number of early-adopters (end users).
The result has been a stellar evolution rate since only last November.
That said, yes, it is 'only' "one step closer," but I agree with your assessment that it is yet again a...
@HistoPol @ruialves What is not covered by either this article or most other discourses, is that ChatGPT isn’t one-directional. Immediate access to the internet will close the loop and it’s own outputs will start to influence itself.
It already was, because it’s not just the technology that forms the network. It’s the humans. We are the other half of the hypercomputer this whole system becomes.
@HistoPol @ruialves Social media companies have already proved uncapable of controllig the products products they maintain. The networks are already too complex to control.
I don’t think OpenAI is even trying. Seems that it’s a leap of faith by now. We are going to have these networks control larger and larger sections of our life. And nobody’s going to have control, because it’s going to be everywhere all at once.
💯 %.
And, as I have outlined in another thread, "they", as the appropriate pronoun should be 😉, learn all at once.
Important point, too, covered here:
"#Google and #Microsoft’s #chatbots are already #citing one another in a #misinformation shitshow [#enshittificatiom] / #Microsoft’s #Bing said Google’s #Bard had been shut down after it misread a story citing a tweet sourced from a joke. It’s not a good sign for the future of online misinformation."
https://www.theverge.com/2023/3/22/23651564/google-microsoft-bard-bing-chatbots-misinformation
(5/n)
...a "giant leap," as #NeilArmstrong phrased it, but not "for mankind," but towards the abyss.
Why? Well it is not just the almost complete access to the wealth of freely available information on a quasi global scale, as well as the network effects that come with it (e.g. just comparing and translating all #Wikipedia entries into and from all languages), but even more worrisome: the loss of a time delay. This moves removes a huge obstacle towards #LLM's...
(6/n)
...developing from #GAI's into #AGI', sentient or not: #ChatGPT has arrived in the present, thus removing one of the most important dimensions that separated it from #RL.
Another important barrier has already been broken, #AI has entered the #biosphere, although that tech is in its very early stages:
https://mastodon.social/@WarnerCrocker/110095775021931183
But that might not even be the final frontier for the emergence of real #AGI.
That is combination of #robotics and #AI.
And, you might...
(7/n)
...have guessed it, it is also making big strides, one company being...(Bingo!) yet again, #Microsoft:
https://mastodon.social/@HistoPol/110129405482528991.
I do not want to enter in systems theory and the like here (see link at the end), but while it is nice and correct to call for
"Developers have the responsibility to ensure that #AI systems like #ChatGPT are developed ethically, with safeguards in place to prevent any misuse or harm."
This, I am sorry to say, is wishful thinking...
(8/n)
...we live in a world, in which #Niccolò #Machiavelli's #Prince(s) have the resources of #Croesus:
https://electrek.co/2023/05/16/tesla-bot-sideshow-new-footage-robots/
And though #Elmo is probably the worst incarnation of a #misanthropic #billionaire, #Zuckerberg has only been less successful in his visions (#Metaverse...a, fortunately failed attempt at #TheMatrix), but has been more intelligent in hiding his character than the former.
How is the saying?--Only trust a billionaire as far as you can throw...
(9/n)
...him. Well how far could you throw #Bezos who's been building data centers around the world, collecting information on millions and consumers and literally one of to go-to companies in the #West for globally scalable B2B cloud solutions.
But then, some animals are more evil than others, to borrow a bit from #Orwell again. Let's not forget about a forgotten #SciFi movie, #Wargames, which again--you might have guessed it, by #Elmo's former #DarthVader mentor,...
*PeterThiel: The Godfather of #2034--A Reboot Of #1984*
(10/n)
...the #German-born #Billionaire #PeterThiel 1), who financed #JDVance's election to #Veep and who l thinks 👉we will have an #AI military "general" productized #SaaS soon enough as #AIP, #Palantir's #ArtificialIntelligencePlatform👈:
https://mastodon.social/@HistoPol/110323739545391429.
And let's not forget, the many feats #Thiel has already accomplished with the military-grade #Spyware, #Palantir, from...
1) * 1967 in #Frankfurt a. M.
*PeterThiel:...*
(11/n)
...#RacialProfiling 3.0 for the #LAPD:
"The #ManhattanInstitute (#MI) created the #racist policy “#BrokenWindows” in the 1990s which was popularized by racist Mayor #RudyGiuliani. It involved “#PreemptivePolicing.” [a potential precursor to the #dystopian #MinorityReport.]
At MI, #PeterThiel took this idea, worked with the #NYPD, and translated it into #Palantir, one of the 👉most dangerous #surveillance and #psyops companies in the world...👈"
Attached: 1 image The Manhattan Institute (MI) created the racist policy “Broken Windows” in the 1990s which was popularized by racist Mayor Rudy Giuliani. It involved “preemptive policing.” At MI, Peter Thiel took this idea, worked with the NYPD, and translated it into Palantir, one of the most dangerous surveillance and psyops companies in the world which is used by a huge part of the government, military and law enforcement. 8/ https://theintercept.com/2021/01/30/lapd-palantir-data-driven-policing/
*PeterThiel... *
(12/n)
"...which is used by a huge part of the #government, military and law enforcement..."
Or he is siphoning off highly personal #medical information from the #UK's #NHS hospitals:
https://mastodon.social/@HistoPol/110194060167267823
So, when #Thiel is not offering #Palantir's #Data #Dragnet to authoritarian regimes in #SouthAmerica and elsewhere to hound down dissidents and unwanted #journalists, he uses it to "influence" elections in the #EU:
...and elsewhere.
Small...
*...Thiel...*
(13/n)
..wonder, then, that even the #Spyware part of #Palantir already had democratic governments worried:
https://cktn.todon.de/@CKsTechNews/109892128985385003.
I will not even begin to talk about how #Thiel and his #TechBros have been working to dethrone the #USD as the world's reserve currency, and would probably be delighted if the #US defaulted.
The fish stinks from the head down. Who really thinks that end-of-the-corporate-food-chain developers will have any say about safeguards and ethics in #AI...
*PeterThiel...*
(14/n)
..companies. He may speak up now or hold his truth forever.
Ok, back to #RL. This is what happened to #Google's #AI developers when they raised the alarm at the erstwhile "Don't be evil" company:
https://jawns.club/@clayote/110308555235252935
And 👉only weeks later, even #Google's #Godfather of #AI,
Dr. #GeoffreyHinton, resigns👈:
https://mastodon.social/@HistoPol/110300734884059290.
So, no, I have hardly any faith that, given the #tech industry having unequivocally found its #NextBigThing, and the market expected...
@[email protected] Assuming you're sincere, here's the paper that warned Google about AI, which resulted in the authors getting fired from Google https://dl.acm.org/doi/pdf/10.1145/3442188.3445922
*PeterThiel...*
(15/16)
...to show explosive growth rates e.g.
https://www.forbes.com/advisor/business/ai-statistics/.
In an interim conclusion, as the English politician and #Historian, 👉#LordActon's famously said:
"#Power tends to #corrupt and absolute power corrupts absolutely.👈"
"Great men are almost always bad men, even when they exercise influence and not authority;
still more when you superadd the tendency of the certainty of corruption by authority.”*
And, as we all know, money...
*PeterThiel: The Godfather of #2034--A Reboot Of #1984*
(16/17)
...can't buy you love, but it most definitely can buy you power.
Oh, a semi-final thing, this is how #Google's big boss approaches the subjects of #AI security and ethics:
https://mastodon.social/@HistoPol/110218725158345991
i/:The prosecution rests its case, Your Honor./i
And, if you are not quite convinced yet and...
(17/17)
...want to have a glance at the #SystemDynamics, #robotics , #AIEthics, among other things, aspects of these #GAI's, you might like to read on here:
@ruialves
PS:
Here is the missing podcast link (2nd article, ca. 11:25 min):
"Artificial intelligence and jobs--beyond the hype"
The Economist Podcasts: Editor’s Picks: May 15th 2023
via @getSonnet
Rui Alves
HistoPol (#HP) 🏴 🇺🇸 🏴
Toni Aittoniemi
