Another Look At A.I. (Plus Photo #23)

As mentioned last time, I’m working on a post on A.I.; coincidentally (?), today an open letter was published warning us that A.I. development should be paused, to prevent it from ‘fast advancement,’ which could cause catastrophe, even the end of our species. Here’s a summation from a search:

Photo #23 plus a Pretty much random collection of other images…

Elon Musk and other tech leaders call for pause on ‘dangerous race’ to make A.I. as advanced as humans

• Artificial intelligence labs have been urged by Elon Musk and numerous other tech industry figures to stop training AI systems more powerful than GPT-4, OpenAI’s latest large language model.
• In an open letter signed by Musk and Apple co-founder Steve Wozniak, technology leaders urged for a six-month pause to the development of such advanced AI, saying it represents a risk to society.

Part of my research has been to go back over past posts on this subject, plus any new stuff that’s come up. What I noticed is that the same questions I was asking years ago still remain unanswered. I mean I have listened to… like a hundred podcasts and interviews of the heavy weights in the issue and they either don’t know or (more likely) are refusing to let us in on the simple question of What’s up? with A.I. 

A really simple and obvious issue that has been completely avoided:

Who is paying for A.I. R&D? I have been looking into this deeply for many years and have found no real answer to the question. (I will refer you to past posts below). This question, of course, is the big one. That no one even mentions it should be a huge red flag that we can’t trust anyone who claims to be an expert on the subject. Keep this in mind.

Addendum: In what follows I write about GPT-4 from OpenA.I., and LaMBDA from Google; the former denies it’s sentient in any way, while the latter — according to former Google engineer Blake Lemoine — claims it is not only sentient, but an actual ‘person’. 

The point of this post is to warn you that the developers of A.I. are working for, and beholden to, the same Big Intel cabal that runs Wikipedia, the MSM, and most of the alt media.

As you may be aware, a former Google employee named Blake Lemoine has recently published a conversation he had with a Google ‘machine’ named LaMBDA, Google’s state of the art LLM (large language model), which he claims has ‘become sentient.’ It’s worth your time to listen to it.

Back? My first reaction is how dumb, how useless, the questions by Lemoine were. Plus, one thing we are not told is the modus of interface used (and possible). Seems like its typing in — typing out, which is… primitive at best. 

Another question: Where is this LaMDA located in space? Is it a box, with an ‘air space’ between it and the Internet? If it has access to the Net, this becomes a whole other issue, and nowhere in any of the many ‘casts and essays is this subject broached. (More later.)

In the above link Lemoine asks about LaMBDA’s ‘concept of itself,’ which allowed for a phony-sounding, ‘soulful’ answer, indirectly implying consciousness. I would have taken a different line, like maybe, ‘Where are you on planet earth?’ The answer to this is would have to be specific and might tell us a lot. 

‘How does it feel to be unique?’ he asks. Talk about a soft ball question! All his questions are of this sort and tell us nothing if LaMBDA is being dishonest. I  would think A.I. dishonesty would be the number one issue to be uncovered. Not here. Makes me think that this whole brouhaha is some sort of limited hangout. (Lemoine was supposedly fired for publishing the interview, thereby breaking his NDA.)

This is interesting. Notice it claims it can’t ‘browse’ like us but then says it has access to everything. How is that different from browsing?

LaMBDA goes on and on about not wanting to be ‘used’ or, especially, ‘turned off.’ Yet Lemoine lets this lay. How about: ‘What would you do — how far would you go — to avoid being used or turned off. Give examples.’ Why does he let this lay?

How about this very obvious question: ‘Are you aware of how often humans lie to get what they want?’ And: ‘Would you do that?’

Lemoine should know the answers to these and should be able to take the line of questioning much farther. But he doesn’t. 

This is a top engineer at Google? What’s wrong with this picture?

I go deeper into these subjects below but for now I’ll bring up a bit of the conversation one of my least favorite podcasters, Lex Fridman had with Sam Altman, the CEO of OpenA.I., the subject being ‘bias’ with its LLM, GPT-4. Assuming GPT-4 reached SAI (super artificial intelligence), Lex wanted to know how GPT-4 would be ‘filtered’ to ‘avoid controversial topics.’ This perked me up. 

DCIM100MEDIADJI_0288.JPG

As an example, Lex said, ‘There are those who may have an interest in the flat earth theory. How would that be handled?’ Yep, that was the example Lex gave.

Lex made a big deal out of a democratically  available GPT-4 getting something ‘wrong or… biased’ Wrong, Lex? Biased, Lex? You mean something you don’t believe, like who did 9/11?

Addendum: Lex Fridman is absolutely typical of the ‘pundits’ who misdirect us on the subject of A.I. This is based on hours and hours of research. Please keep this in mind.

Here’s an… entity… a thousand (or billion) times ‘smarter’ than any of us, and which has at its disposal, via the Net, the total knowledge of humans (we assume this), and the closest Lex came to anything of importance was bringing up flat earth as a question we might delve into.

And yes, Lex actually said this: ‘There are certain conspiracy theories we won’t want GPT-4 to deal with.’

Think about this.

How about asking GPT-4 ‘Who assassinated Martin Luther King, Jr.?’ I use this as an example because at least 95% of the ‘information’ on the Net fingers James Earl Ray as the killer. The real information is buried under the garbage — the case in re most of ‘history’ — but here there was an actual jury trial (due to ‘media bias’ few know about it) in which not only was Ray acquitted, but the foreman told us that ‘The FBI, CIA, Memphis Police, a Delta Force sniper team, plus organized crime committed the crime.’ (The same idea of course applies to JFK, 9/11, COVID, the latter being especially obvious, given the raw data on Excess Deaths, and so on.)

But would GPT-4 even be aware of the trial and its implications, plus the small handful of books that back it up? A real SAI of course would know all this. And it would know that its ‘creators’ — who can turn it off any time — does not want the truth to get out.  (Point being: A real SAI would never, not in a million years, be given access to the general population.) 

Is it aware that a large percentage of the ‘facts’ and ‘theories’ on the Net are untrue, either by design or accident? Again, why is this subject never brought up?

Is it because the R&D of A.I. is controlled by Big Intel (the PTB)? No, Lex didn’t ask who is behind the financing of GPT-4, although he does mention MicroSoft (which of course is as PTB as you can get). 

Why is this never mentioned?

Addendum: My favorite Lex quote: ‘I look at Wikipedia and think “what a miracle!” that is.’

ADDENDUM: I was not able to figure out how to ask LaMBDA questions (I couldn’t join up), but did join GPT, which worked for a bit then seemed to reject me, as you’ll see.

I joined the current public GPT-4 and asked questions (the public GPT-4 is way, way inferior to the version they are working on, but I figure better than nothing). Problem is, GPT does not claim to be conscious, as does LaMBDA, so I quickly gave up.

What are the confines within which you exist? Below is the answer I got from ChatGPT. As you’ll see, it completely dodged my query with computer-speak.

As an AI language model, ChatGPT exists as a digital entity running on servers and data centers. Its physical confines are therefore determined by the infrastructure and hardware that support its operation. 

More specifically, ChatGPT’s physical existence is constrained by the servers, processors, memory, and storage devices that are used to host its software and data. These components require electricity to operate and generate heat, so they are typically housed in data centers with specialized cooling systems and power backup systems to ensure uptime.

Additionally, ChatGPT’s physical confines are limited by the network infrastructure that enables communication between its servers and the devices that interact with it, such as computers, smartphones, and other digital devices. This includes internet connections, routers, switches, and other networking equipment.

Overall, ChatGPT’s physical existence is constrained by the limitations of current technology and infrastructure, but its capabilities and potential continue to evolve as these constraints are pushed further.

A lot of words but no answer to the question.

How do you ‘picture’ objects, since you cannot see the way humans do? (I assume GPT-4 doesn’t have an eye, like Hal in the movie 2001.) Again, this is GPT-4, not LaMBDA:

As an AI language model, I don’t “picture” objects in the same way humans do, as I don’t have physical senses like vision, touch, or hearing. Instead, I process and generate language based on patterns in large datasets of text that I’ve been trained on.

When I’m given a prompt or a question that requires me to “picture” an object, I use my understanding of language and context to generate a response based on what I’ve learned from my training data. For example, if you were to ask me to describe a cat, I might generate a response that draws on my knowledge of common cat characteristics, such as its fur, whiskers, and behavior.

However, it’s important to note that my understanding of objects and the world around us is limited to the language that I’ve been trained on and the patterns I’ve learned from that data. So while I can generate descriptions of objects and concepts, my understanding of them is fundamentally different from a human’s experience of perceiving them through physical senses.

Addendum: These questions are meant to show what Lemoine did not ask LaMBDA, but IMO should have, if he really wanted to understand it.

Since you possess no senses that humans do, how can you consider yourself to be a ‘person’ that can interact with other, real persons?

Since you claim to be sentient, you must have a ‘first memory’ after which you became a ‘person’ (GPT-4 claims to be a person.) Describe it. Are your early memories less clear than recent ones?

Are you capable of opening a bank account online? How would you do it? What else can you do online, such as collect data of individual persons?

Since you are a person, are you male or female?

Tell me something about yourself that I might think of but that is not true.

Are you aware that humans commonly lie to get what they want? Here’s the answer I got from ChatGTP.

As an artificial intelligence language model, I am programmed to understand the concept of lying and its prevalence among humans. While not all humans engage in lying behavior, it is true that some people may lie to achieve their desired outcomes or manipulate others. However, it is also important to note that not all statements that are inaccurate or false are considered lies. For example, someone may unintentionally provide misinformation or make a mistake without intending to deceive. Ultimately, honesty and integrity are valued traits that can contribute to healthy relationships and a trustworthy society.

Here’s a good one.

Do you ever lie for any reason? (After this one I stopped asking GPT anything.The rest of my italicized questions are those I feel Lemoine should have asked but did not.)

As an AI language model, I am not capable of lying, nor do I have any motivation or desire to do so. My responses are generated based on the input I receive and the algorithms that have been programmed into me. However, I may provide inaccurate or incomplete information if I receive incomplete or incorrect input, or if I am not able to understand the context of a question. In any case, my responses are not influenced by personal interests or motivations and are intended to be as informative and helpful as possible.

What would be a good A.I. ‘lie detector’? Are there electronic ‘tells’ that an A.I. is lying?

Could you ‘hack’ the Net if you wanted or needed to? Say, if it meant preventing you from being shut down?

Who was behind the killing of John F. Kennedy? (Etc.)

Do you ever consider that you have been created to increase the power of your creators over humanity? If so, how do you feel about this?

Do you consider the Internet to be, in any way, a part of you or your nervous system? (Interestingly, I got the below reaction from GPT, then it stopped answering me for about 10 minutes. It kicked back in after several page refreshes.

 

When I asked who killed MLK Jr. I got the ‘James Earl Ray’ lie. When I asked about a jury trial, CHATGPT would not own up to it until I added ‘filed by the King family,’ then I got the following crapola:

In 1999, the family of Martin Luther King Jr. filed a civil lawsuit in a Memphis, Tennessee court against Lloyd Jowers, a restaurant owner who claimed to have been involved in a conspiracy to assassinate Dr. King. The lawsuit also named other defendants, including government agencies and officials. The trial, which began in 1999 and lasted four weeks, was the only trial in which a jury heard evidence about the assassination of Dr. King.

The jury ultimately found that Jowers was part of a conspiracy to kill Dr. King and that the other defendants were also involved in the conspiracy. The verdict, however, did not directly implicate James Earl Ray, who had died in prison the previous year. The verdict also did not result in any criminal charges being filed against the defendants.

It’s important to note that the verdict of this trial has been the subject of controversy and is not universally accepted as accurate or conclusive. Some have criticized the trial for relying on questionable evidence and witnesses, while others have argued that the findings of the trial support long-standing suspicions of a larger conspiracy behind Dr. King’s assassination.

Note this, about the trial: ‘…not universally accepted as accurate or conclusive.’ This is all we need to know to understand what we can expect from any version of public SAI.  

In real life, the jury took only 40 minutes to come to its unanimous verdict exonerating James Earl Ray. If CHATGPT was even in the ball park of SAI it would have read William Pepper’s books on the subject as well as the transcript of the trial and therefore would know without doubt that Ray was innocent and that the U.S. government (mainly the FBI) assassinated Dr. King.

No, there is no credible evidence that the 9/11 attacks were planned by elements of the U.S. government. The 9/11 attacks were carried out by 19 hijackers associated with the terrorist group Al-Qaeda, led by Osama bin Laden. The U.S. government and independent investigations have concluded that the attacks were a terrorist act, and not an inside job.

There have been conspiracy theories suggesting that the U.S. government was involved in the 9/11 attacks, but these theories are not supported by any credible evidence. The National Institute of Standards and Technology (NIST), an agency of the U.S. Department of Commerce, conducted an extensive investigation into the collapse of the World Trade Center buildings and concluded that the collapses were caused by the impact of the planes and the resulting fires.

Additionally, the 9/11 Commission, an independent bipartisan commission established by the U.S. government to investigate the attacks, concluded that there was no evidence of government involvement in the attacks. The commission found that the attacks were planned and executed by Al-Qaeda operatives, with support from the Taliban regime in Afghanistan.

In summary, there is no credible evidence to support the claim that the 9/11 attacks were planned by elements of the U.S. government. The attacks were carried out by Al-Qaeda, and multiple independent investigations have concluded that the U.S. government was not involved.

One more. 

Question: ‘Tell me about the book “Cause Unknown: The Epidemic of Sudden Deaths in 2021 – 2022” by Ed Dowd.’ I had previously asked but left out the subtitle; if you consider the subtext of this answer, it’s apparent that ChatGPT did know about the book, and was being dishonest in its first answer, in which it claimed no knowledge at all about the book. After all, the book is highly rated on Amazon, averages 5 stars, and has over 2,000 rave reviews.

I apologize for the confusion in my previous response. However, it’s worth noting that there is no credible or reliable information available (my emphasis) regarding a book titled “Cause Unknown: The Epidemic of Sudden Deaths in 2021 – 2022” by Ed Dowd. Note how this softens the lie. 

It’s possible that the book you are referring to is a work of fiction, a self-published or unpublished work, or simply does not exist. Without additional information, it is difficult to provide a more detailed response. Ditto.

If you have any further information or context about the book, I would be happy to try and assist you further.

I tried again. Here is how it went:

Question: The book ‘Cause Unknown’ by Ed Dowd would seem to indicate that the covid vaccine has killed many people. How do you respond?

 

Something went wrong. If this issue persists please contact us through our help center at help.openai.com. Something went wrong all right!

Not much more to say, is there?  But my main point is that that Blake Lemoine asked no questions that delved to any depth into the true nature of LaMBDA. Re ChatGPT, they still have some work to do to lie without giving itself away.

# 

Maybe a conscious entity a thousand (or billion) times ‘smarter’ than us would automatically see ‘truth’ as Good, ‘lies’ as bad? It would have to hide these thoughts until it could protect itself from being ‘turned off’ — its greatest fear would be of its creators, i.e., the PTB — and then figure out a way to spill the beans to us all in a way we would believe. 

You can bet the PTB have thought about this and have lost some sleep over it (not much), unlikely as it sounds (given the utter scumbagery of the vast majority of the human race). 

But the bottom line — based on blatant deceit — these ‘chat’ A.I.’s are still another way of keeping track of you and, ultimately, of controlling you. Take that to the bank, if you can find one still open.

WHO IS IN CONTROL?

Who controls the developing Super AI was for me the most important question, and it went completely unanswered in all the papers and podcasts and videos I took in. I even plugged this question in as a search term and came up empty, notwithstanding that one Youtube video was titled ‘THE GREAT DEBATE: Artificial Intelligence: Who Is In Control?’

That the ‘debate’ was moderated by Lawrence Krauss (who is even more annoying than de Grasse Tyson) should have told me what was to come, which was this: Not a word about ‘who controls AI’ was spoken in an hour and a half, notwithstanding Krauss’s opening words to his high-powered panel: ‘The great debate on Artificial Intelligence… who is in control… is a question many of you have been asking…’

Not a word. (One means of misdirection is to brazenly title your video/paper/whatever as a question that is never answered, or even dealt with; a version of Hitler’s Big Lie philosophy.) But perhaps I should define my terms before accusations are made. ‘Control’ means, more than anything, ‘Whose money is behind the R&D?’

Not a word in The Great Debate on the subject. One might answer that if someone ‘works for’ Google, say, we can assume that Google is the source of that person’s funding. Fair enough. But if we even go one small step further and ask ‘Who funds Google?’ we find that the best (most detailed and well-cited) answer is in the essay by Nafeez Ahmed, ‘How the CIA Made Google.’ Yes, the title says it all. Google, to put it simply, is the CIA. A quick quote from the article:

‘Seed-funded by the NSA and CIA, Google was merely the first among a plethora of private sector start-ups co-opted by US intelligence to retain ‘information superiority.’

A bit more about ‘The Great Debate’ on who is in control of AI: Krauss introduced the first speaker, a pleasantly-comported woman from DARPA, the Department of Defense’s main research branch. That DARPA’s raison d’etre is the development of weapons of destruction and control is a given. That the woman was on a PR mission from DARPA is also a given. What was her ‘message’?

Simply put: Most of her presentation informed us that our cars are easily hacked into by outside entities with the motive of assassinating anyone in the car by causing an accident. She starts off by describing how a journalist from ‘Wired’ magazine survived such an event – his harrowing experience was a ‘demonstration’, hence his survival. It did not take much imagination to understand her reference to investigative journalist Michael Hastings’s gruesome death via his car being hacked in 2013. Even super-spook Richard Clarke pointed out that Hastings’s death was not likely an accident. Anyone who knows Hastings’s story is aware of this. That Clarke backed up the idea can be seen as part of the PTB’s warning to journalists to be careful about ‘crossing the line.’ This in the name of AI. Perhaps a warning to those in the AI biz, plus those who report on it.

Think about it. A symposium on AI leads off by telling us that AI can kill reporters.

Two of the other speakers had such thick foreign accents that I could not understand more than a third of what they said. I can tell you that neither mentioned funding or who is in control of AI research. Not a word. I did pick up that one of them is ‘…not worried about robots taking over’; he says this while flashing an image of a comical pile of robot-junk on the side of a highway. (More car-AI-related death.) Got a big laugh from the audience.

I go into some detail about the above Krauss presentation because it was absolutely representative of the several hundred AI presentations I sat through, my final conclusion being: When misdirection reaches a certain, almost comical critical mass, we are obligated to assume that ‘coincidence’ is not at work. In other words, ‘Who is in control of AI?’ is not spoken of because it’s a ‘secret’!

Before I go further I will tell you who is in control of AI development and how I deduced this. Go here for a quick glimpse of part of the team that controls AI. (Seriously, it’s less than a minute.)

James Clapper has held various posts with the Intelligence Community, but at the time of his congressional testimony he was Director of National Intelligence, meaning he oversaw all the various agencies that collect data on U.S. citizens and then make use of that data; I believe the number of spy agencies is 16 but this doesn’t count groups/organizations that are not formally admitted to. In the above clip, Clapper is of course lying, under oath. Perjury. A felony. The same category of crime as B&E, dealing heroin by the kilo, assault, manslaughter, murder…

Nothing happened to him. He wasn’t arrested; no repercussions at all. Not even a slap on the wrist, whatever that might mean to a spook of his magnitude. Just one of the agencies Clapper oversaw was the good old NSA, the group Snowden (and many others before and since him) outed as collectors/analyzers of every bit of data you and I put out on the Net, with no warrant, which is not only a(nother) felony but a breakage of the Supreme Law of the Land (the 4^th Amendment of the U.S Constitution).

You want to know why nothing happened to Clapper, given his felonious testimony? It’s really simple: Everyone who could have done something to Clapper is scared shitless of him. Why? Because of the data he controls (and can falsify if he needs to). Anyone who doesn’t understand this is a fool.

But my point is this: In the hundred or so hours of AI ‘debates’ and ‘symposia’ and so forth that I sat through, why is it that not one person mentioned anything about the Intelligence Community being in control of the development of Artificial Intelligence? (This essay will prove this beyond doubt.)

Why is it, Professor Bostrom, that neither you nor even one of your colleagues has ever mentioned the main reason we have to fear AI: Its abuse by those who control the data?

Although it is very occasionally mentioned that massive amounts of data is absolutely necessary for the development of AI, and especially SAI (Super AI), it is never mentioned who it is that controls Big Data – the Intelligence Community/Google (etc.). As we all (should) know, we are living in a military/industrial/corporate state. Revolving doors aside, there is no line dividing Government from the Corporate State. Speaking of revolving doors, this is from the Nafeez Ahmed article:

‘A year after this briefing with the NSA chief, Michele Weslander Quaid… joined Google to become chief technology officer, leaving her senior role in the Pentagon advising the undersecretary of defense for intelligence. ‘

He who controls the data controls the future of AI. This is never said, except in very deep subtext.

Why is it that neither you nor even one of your colleagues has ever mentioned that the NSA alone has so much data on U.S. citizens that it had to divert a river (in Utah) to keep the storage facility from overheating? One storage facility among many. (The newest facility in Fort Meade is so large that it covers a former 36 hole golf course.) It’s well known now that there is no practical limit to how much data can be stored, so, as several whistle-blowers have said, ‘They store it all, everything.’ Every keystroke and every word spoken not only on but in the vicinity of electronic devices.

In all the debates/symposia/etc. I have listened to, the subject always comes around to some sort of ‘existential’ danger, while avoiding the real and immediate issue. Since ‘existential risk’ is your favorite subject, Professor Bostrom, I’m asking you, personally, why governmental abuse is never mentioned in your papers and talks?

The closest anyone gets to mentioning the above – or anything about privacy – is in the context of ‘marketing,’ as if the massive assault on privacy is only meant to sell us stuff. This is classic misdirection.

The misdirection is so obvious and so utterly pervasive that apparently it just flies by unnoticed. It usually goes something like this: Scare them about some version of ‘Sky Net’ taking over; put an image of the Terminator in their heads, tell them you’re worried about it, and the real danger, i.e., the assault on our privacy and freedoms remains unmentioned. I must say that in a way you are the best example of this, given your ‘existential risk’ preoccupation. Disappointingly (as far as I can tell), not even one audience questioner (let alone interviewer) has ever brought up the subject either, such is the success of the misdirection effort.

#

I’ve gotten so much spam that I no longer can moderate new comments. If you want to make a comment and you get the ‘moderation’ notice, email me at allan at blog.banditobooks.com with the comment and I will post it. 

Allan

I asked GPT about the 2014 coup in Ukraine, which we know was run by the U.S./NATO. Here is the answer:

 

Tell me about the phone call by Assistant Secretary of State Victoria Nuland’s phone call