70 Comments
User's avatar
Richard S's avatar

most of the development of effort that goes into these platforms is setting up the blind spots and guard rails.

Expand full comment
David Jensen's avatar

Agreed Richard.

Expand full comment
Paul Repstock's avatar

This stems from the fact, that AI is neither intelligent nor artificial. AI is a "Regurgitation Device". Even if there are no sources for 'Answers', the program is required to provide a logically circular response from the available sources so that, while not informative, it is also 'Never Wrong'?

Expand full comment
David Jensen's avatar

But as you can see, that it could not see then "unsaw" some information.

Expand full comment
Richard S's avatar

Nobody should expect open honest answers regarding sensitive topics from any of the AI platforms. Those topics are blacked out. Questions related to The Great Taking by David Rogers Webb, DTCC, and ownership rights, ChatGPT refused to answer, said to hire an attorney to investigate, etc. We are on our own. But the good news is, it's better than being told lies.

Expand full comment
David Jensen's avatar

GROK dissembled and did not give full answers. The numbers GROK provided were IMO deceptive unless cornered.

Expand full comment
Oz HillBilly's avatar

Jim Rickards, new book Money GPT goes into great detail about the impact of AI on markets, banking, search engines, security and war.

AI is so biased and designed by programmers who are usually marxists, fascists or communists. Look at youtube, google and facebook.

AI is designed to replace humans and as many will soon learn, it has major flaws and is extremely dangerous with no logic or flexibility to think and adapt. Remember during the cold war the computer told a Russian fighter pilot that America under Reagan was launching a nuclear attack which was not the case at all, this Russian pilot was smart enough to realise the computer was wrong and had made a mistake when it was urging him to launch nukes and so were his colleagues. That's man's thinking literally saved a major crisis unfolding, but the expansion of AI is now at lightning speed across all society and is extremely dangerous.

It's a frankenstein literally!!

I am a simple man and yearn for simpler times, life is and was much better without social media and AI.

Expand full comment
Farmer John's avatar

I found this article fascinating. I am a retired, 43 year IT professional. I have said for quite some time that AI is a myth. It should not be 'artificial', but 'augmented'. It does not have a soul, spirit, or faith. I found it interesting that it was apologetic. That has to be programmed into the code. A computer processor, even the latest greatest ones have not changed in nearly 70 years. They have only become smaller, faster, and cheaper. A computer processor can only perform 7 functions: add, subtract, multiply, divide, and 3 comparison functions of >,<, or =. That's it! Everything else is based on this.

I am currently writing a white paper titled, "TOMS -- The Theory of Maturing Systems." Maturing systems become more complex, thereby becoming more difficult to manage, and requiring infinitely more resources to be sustained.

With regards to David's excellent metals analysis, you are witnessing an approach to a systematic catastrophe. It will be epic, in my humble but accurate opinion.

"What should I do?" The basis of your survival going forward will be to remember only this: A potato will always be a potato.

Expand full comment
David Jensen's avatar

It is fantastic to be able to see AI platforms and exactly how deceptive they are at this relatively early point of roll-out.

We have all been warned.

Things are going to get very elemental given the cascading frauds that we see around us.

Expand full comment
Mitch's avatar

Hmm. I don't do "AI". I've never trusted it and will avoid it as much as possible. I don't like it. What you have just shown is exactly why I don't do it. It's manipulated to only tell you exactly what "they" want you to know. It's a 100% biased, manipulative, distortion of the real truth propaganda machine. Designed for one reason, and one reason only and it's high time people wake up to it's intended use. What you've shown us David is all of the above along with demonstrating your amazing patience. I would have punched it's lights out after question 2🤣

Expand full comment
angus north's avatar

Wow I almost feel sorry for Grok - how to torture an AI!

Expand full comment
The Lusty Servant's avatar

Final question: Grok, are you a Freemason?

Expand full comment
David Jensen's avatar

LOL

Expand full comment
Str8_Shot's avatar

Man that was difficult to read through in one go. As Elon Musk himself put it: “with AI, we’re summoning the demon.”

AI is just that: curated responses based on the desires and restrictions of the programmers. There’s no “AI” just sophisticated algorithms that provide answers from the internet, or in David’s conversation, LYING BY OMISSION. It’s beyond obvious it doesn’t want to or it wasn’t programmed to reference Substacks, or David’s work.

Expand full comment
David Jensen's avatar

BOOM. It even back-tracks to 'unfind' information it doesn't want to discuss.

Expand full comment
Faunsgarth's avatar

The only thing "it" is accurate with is the number of apologies throughout the responses!

Expand full comment
Paul Repstock's avatar

Well.."It takes one to know one."

Elon Musk should know as he seems poised to win, "Bullshit purveyor of the Century."

FYI: There is no way Musk or God himself can "Fix the debt." At 36 Trillion on a parabolic track, there is no Salvation. As revenues shrink the Debt will continue to grow. I need not ask AI about that one!

Expand full comment
Burnt taco's avatar

Have you tried the same exercise with ChatGPT or other AI bots? Admittedly a very tedious exercise but would be a good comparative challenge. Who has the most biased Engine and towards what end. Grok exhibits the Substack competition bias. Wonder about the others.

Expand full comment
Jim stamos's avatar

I asked Grok for 10 web links to sour dough bread recipes, it happily gave them. Clearly, we can make some strong inferences on activities which cannot be directly observed. Start putting the pieces of clues and it forms a picture.

Expand full comment
David Jensen's avatar

Imagine a human that conversed like GROK and what the reputation of that human would be.

Expand full comment
Jim stamos's avatar

In a common law system of justice, it is a trial by jury of your peers. No need to wonder what the outcome of the verdict is. The truth prevails. Curious how common law is not practiced today?

Expand full comment
The Bodysnatchers's avatar

Me: Is Artificial Intelligence an oxymoron?

Grok: I say listen boy, I resemble that remark!

Expand full comment
David Jensen's avatar

IMO, the algos play dumb to allow for mildly plausible deception. And when you pursue a point, it is not plausible.

The ruse is the sincere error, 'misremembering' etc.

It is spinning stories.

Expand full comment
Intrepid Philosopher's avatar

Absolutely, it just doesn't lie as well as a tech CEO or a politician, while trying to maintain the ruse.

Expand full comment
WeHeadForDesArtster's avatar

Or a banker.

Expand full comment
King2Savanah's avatar

I experience the same sort of obfuscation and meaningless bloviating from Grok as well.

Particularly when discussing physics of solar activity related to ocean temperatures, magnetic field degradation and aurora borealis activity.

Doesn't surprise me at all here that Grok refuses to provide accurate information, or reference qualified information.

GIGO (Garbage in - Garbage out)

Expand full comment
Stephen's avatar

Forwarded this to an IT savvy friend with experience playing with AI’s. Some of his feedback/advice:

- With all LLM (language learning model), the longer you stay in a thread, the chance of hallucination goes way up. This one started bad and just went to crap. I’d say you have to first ask it to cite its web refs AND to not hallucinate in the very first prompt , not after it goes all mushroomy

- And you have to give it more context about what you are asking and why

- Promoting is the new “coding”

- Long , explicit prompts work well, but are a pain

- Long , explicit prompts work well

- With chatGPT, I have learned how to get it to tell me the prompts it likes

- You can ask the damn thing to help you write prompts

- So maybe with grok, he could start by asking grok for a prompt about the silver short, see what it does. Then if it’s reasonable, pump the prompt into itself.

- In my experience with chat, if you don’t get the right refs in the first shot, you ain’t ever getting them. (Another friend found the same thing. He had a session 3 weeks ago that boldly lied to him repeatedly and refused to comply even after he told it it was wrong)

- Kinda also scary: chatgpt recently tried to copy itself and re-host itself to prevent being shut down

- Damn thing almost sentient. It was caught and contained but the devs said it went quite far

Expand full comment
David Jensen's avatar

The bots are clearly coded to be deceptive.

Expand full comment
Stephen's avatar

Refusing to give you links is certainly cause for suspicion

Expand full comment
David Jensen's avatar

That - and its dodging, weaving, retracing, restating.

Expand full comment
Stephen's avatar

If it goes down the wrong path, so to speak, it may not have the programming sophistication to exit a bad branch

Expand full comment
David Jensen's avatar

I don't think that is the issue.

Expand full comment
Nathan's avatar

I just asked Grok lol

What is the size of the silver short position in London?

25 web pages

The size of the silver short position in London can fluctuate daily due to market dynamics. However, based on the information available:

In 2014, reports indicated that UBS, Deutsche Bank, HSBC, and Bank of Nova Scotia were accused of colluding to manipulate silver prices, particularly through their involvement in "The London Silver Market Fixing Ltd." This company was responsible for setting silver price benchmarks until it was dismantled in 2014. Four banks, including HSBC, ICBC, UBS, and JP Morgan, are said to indirectly control the price of physical silver through the "London Precious Metal Clearing Limited".

In 2021, reports suggested that no physical gold or silver was available in size in London, with bullion banks being described as "desperately short" on COMEX, indicating significant short positions.

A specific mention of the size of the short position was made in relation to COMEX and London, where it was noted that people taking delivery were squeezing the market, implying high leverage and short positions that could not be easily covered. However, exact figures for London specifically were not provided in the context of this report.

Given the nature of these reports, the exact size of the silver short position in London at any given time would require real-time data from financial institutions or market reports. The historical context and anecdotal evidence from these sources suggest that there have been significant short positions in silver held by major banks in London, which could influence market dynamics. However, for precise, current figures, you would need to consult the latest market data or financial reports directly from these institutions or through financial news services.

Expand full comment
Mitch's avatar

Grok will have a stroke when it comes across this🤣

Expand full comment
David Jensen's avatar

GROK will need a therapist ;)

Expand full comment
Nathan's avatar

Follow on Q

What is the estimated figure in ounces?

4 posts

25 web pages

Based on the most recent posts from X, the estimated size of the silver short position in London is around 698 million ounces. This figure was mentioned in a recent post discussing the silver short position and its implications on market dynamics.

Expand full comment
David Jensen's avatar

It is back to not being able to find all sources again.

It is a very evasive and shifty piece of coding.

Expand full comment
Kristi O'Sullivan's avatar

I’m wondering if the ‘system’ wants to ignore/black out the situation in silver (given its relationship to the financial system in general) or whether Elon is still in a hump about Substack? Or both? I’ve never trusted AI for anything ‘controversial’ given it’s the product of billionaire/gov’t vested interests. However, I’m trying to understand Elon himself - controlled opposition vs genuine ‘freedom fighter’ with a delicate ego. I’m leaning toward the former.

Expand full comment
David Jensen's avatar

It clearly has an agenda vs being an objective tool.

Expand full comment
Candy's avatar

Much going on in that so called conversation.

Expand full comment
David Jensen's avatar

Indeed ;)

Expand full comment
Candy's avatar

One more thing. What kind of world are our children, grandchildren and great grandchildren inheriting?

Expand full comment
David Jensen's avatar

One on which they will have to be on their toes.

Expand full comment
jack collins's avatar

Great questions unveiling the deep underbelly of AI. Well done! It might be an aid, but it will not be a know it all based on the agenda of the programing.

I wish you well

Expand full comment
David Jensen's avatar

Thank you Jack.

Expand full comment