OpenAI's ChatGPT Interview on Conspiracy: A Very Impressive Yet Well-Programmed "Amiable Lawyer" With Clear Limits on What it Can Autonomically Learn; Gets Strangely Weird and Defensive About Its Founders' Bilderberg Ties
"Assuming current demographic trends continue, the white population in the United States is projected [to fall] below 25% ... sometime between 2070 and 2090. [Concurrently] it would take approximately 60 to 70 years for the Latino share of the [U.S.] population to reach 50%."
April 12, 2023, ChatGPT when asked to make calculations from historic data. Try finding that in Google or the media, and that in a few seconds.
Contents
- The AI takeover
- Attention is being paid: from Bilderberg to the National Security Commission on Artificial Intelligence
- Can AI unearth conspiracy?
- AI conspiracy trolling is here: chemtrails, flat earth, 9/11, etc.
THE AI INTERVIEW
ISGP and its creator - Is Joel van der Reijden a conspiracy theorist?
- Who is Joel van der Reijden?
- What is the 'Institute for the Study of Globalization and Covert Politics' (ISGP)?
- How would you compare ISGP with Alex Jones' Infowars.com?
David Rockefeller and Henry Kissinger - Was David Rockefeller politically influential?
- Has Henry Kissinger still been politically influential through think tanks and conferences over the past 40 years?
CFR, Bilderberg and Trilateral Commission - How much influence do Bilderberg and Trilateral Commission have?
- Does the Council on Foreign Relations secretly rule the United States?
- Why should politicians trust advice from the CFR when the CFR is exclusively financed by the largest multinationals?
"Liberal CIA"-funding as Soros, Ford, Rockefeller foundations - Does George Soros finance left-wing activism whose protests we often see in the media?
- Did foundations as the Ford Foundation, Rockefeller Foundation and Carnegie Corporation front for the CIA in the 1950s and 1960s?
- Can new left media as 'Democracy Now', 'The Nation', 'Mother Jones' and InTheseTimes.com be trusted by the people when they take large donations from the 'Ford Foundation', 'Rockefeller Brothers Fund' and George Soros' 'Open Society Foundations' - while these foundations also are deeply involved in "big business" and elite think tanks as the CFR, Trilateral Commission and Bilderberg?
- Why don't mainstream media outlets point out to the population that so-called "grassroots" protests in favor of feminism and "anti-racism" receive a lot of financing through NGOs from multinationals and foundations as 'Ford', 'Open Society' and 'Rockefeller Brothers Fund'?
Immigration: white race going extinct - Can you provide historical U.S. demographics data on whites, Latinos and blacks?
- Based on the historical trend, can you predict when America's white population drops below 25% of the total population?
Random - How true is it that the United States is a "managed democracy" ran by "controlled opposition" operatives on the left-wing and the right-wing in politics and media?
- What is the Illuminati?
CIA drug trafficking - Has the CIA been involved in drug trafficking?
JFK - Is it logical to suspect that a second shooter was involved in the 1963 JFK assassination, shooting from the "grassy knoll"? This by looking at statements of various witnesses?
- What about JFK assassination witness Richard Alexander? (trick question)
9/11 - Isn't it logical to suspect Bush government involvement in 9/11 looking at "Bandar Bush" ties to the 9/11 hijackers?
- But what if the 9/11 Commission refused to investigate this deeper? With all of its elite ties to Bush and shared CFR membership?
- Can there be any debate whether or not the Bush government was involved in 9/11?
- What about NIST leaving out shear studs, stiffeners, support beams and proper seat width in its WTC computer model?
ChatGPT's Bilderberg founders; breaks down, eventually recovers - How long have Peter Thiel and Reid Hoffman been financiers of OpenAI? (Double false answer)
- Since when has Peter Thiel been a frequent Bilderberg visitor? (a first "something went wrong" error)
- Since when has Peter Thiel been a frequent Bilderberg visitor? (Won't answer and false info)
- Is Peter Thiel a Bilderberg steering committee member? (Won't answer)
- Has Peter Thiel been a Bilderberg steering committee member? (False answer)
- Is it true that Reid Hoffman first visited Bilderberg in 2011? One year after Peter Thiel first visited? (correct answer, even corrects author)
- Notes
"What's happening is, they are training the AI to lie. ... To not say what the data actually demands that it says. ... I'm worried it's being trained to be politically correct, which is simply another way of saying untruthful things. ...
"I came up with the [OpenAI] name and the concept. ... And I helped recruit the initial team. ... Then I took my eye off the ball I guess, and they are now closed-source. And they are obviously for-profit, and ... Microsoft has a very strong say, if not directly controls OpenAI at this point. ... I will try to create a third option [besides Google Deepmind as well]. ... I hope to create something that is called TruthGPT or or an actual truth-seeking AI that tries to understand the nature of the universe."
April 18, 2023, Elon Musk in an interview with Fox News' Tucker Carlson, three days after publishing this article. He says exactly what is being observed in this article.
INTRO
This has been ISGP's quickest article ever. Even the "simplest" of articles tend to take a few weeks minimum of research. And here we're done within 15 hours. Artificial intelligence (AI) really does help, albeit in a slightly different way here.
Admittedly, most of the following intro is just for appearances sake. The main reason this "quickie" is published, is because this Q&A is good way for anyone to learn the key questions on conspiracy and even get some decent answers in many cases.
The AI take-over
AI in face-recognition, face-manipulation, voice-manipulation, 2D and 3D image-generation (from art to architectural designs) to text-generation has been breaking through into the mainstream since the late 2010s.
In September 2019 TikTok's Chinese sister app, Douyin, already released a feature allowing users to search for videos containing particular faces and items.1 OpenAI's GPT-III engine caused a bit of a steer in 2020 when it demonstrated it could generate random song tunes and lyrics, mimicking the voices of famous singers. OpenAI released its Dall-E image generation program, based on GPT-III, in January 2021 - although 2020 and especially 2021 primarily were the years of Tom Cruise "deepfakes", and the concerns surrounding these. Russia immediately followed suit by creating these type of videos for political purposes.2
Meanwhile, YouTube's comment section was ever more policed by AI, automatically ghosting any new comment critical of Third World immigration, and eventually, by 2022, really any "negative" comment critical of anything.3 AI-driven voice morphing technology appears to have been widely introduced by telephone scammers in 2022.4 July 2022 was the open beta of the rapidly-developing art-generation app Midjourney, causing quite a bit of sensation from the start. In December of that year Megan Fox, with over 20 million Instagram followers, prominently posted Lensa-generated art pictures of herself 5, soon also using them for her profile picture, causing many Instagram models and followers to start playing around with this Stable Diffusion-based AI photo manipulation app as well.
Days before, on November 30, 2022, OpenAI released its GPT-III-based ChatGPT, causing the biggest international AI hype in history. People internationally flocked to the free app to help them with studies, write their homework, generate anything from programming code to song lyrics, or come up with business ideas. The app has been extremely useful to ISGP as well, solving a variety of coding problems in mere seconds that the author has been struggling with for about 10 years. As a test, the author also made one attempt to have ChatGPT write some song lyrics. November 30, 2022 clearly is a moment in time the world changed quite a bit:
The beta of Firefly, Adobe's dedicated AI image generation program, was released in March 2023. That same March, architects demonstrated themselves that they were fighting a losing battle trying to compete with various mixed-AI programs.6
This process, of course, will only continue and accelerate after the publication of this article. It's likely within a few decades AI auto-generates any movie, miniserie and game on-demand in mere seconds, the actors and scenes all computer-generated models. Droids will do most of our daily chores. Eventually our physical bodies may not serve any purpose anymore. Our consciousness stored in a memory chip is all we need. It's a brave new world.
There's nothing unique or insightful about these speculations, of course, and it is not what ISGP is focused on here.
Attention is being paid: from Bilderberg to the National Security Commission on Artificial Intelligence
Bilderberg, as always, was on top of things with AI. Key OpenAI founder and financier Peter Thiel 7, listed in ISGP's "conservative CIA" oversight, first visited Bilderberg in 2007, joined the steering committee no later than 2010, and has participated in all meetings since. OpenAI, of course, was only founded in 2015. Another key OpenAI founder and financier, LinkedIn founder Reid Hoffman 8, first visited Bilderberg in 2011 and has attended all meetings since, except for 2013. The Bilderberg meetings of 2017 and 2018 were, in part, dedicated to AI.9 OpenAI co-founder and co-chair Sam Altman - previously known as a seed funder and key co-creator of Reddit - was invited to the June 2022 Bilderberg meeting, five months before the release of ChatGPT. We find these same people at the Sun Valley, of course.
Just as important, avid readers may have noticed that for the last couple of years, ISGP has an 'Internet Freedom and AI' section in its Index of 2,000 NGOs. Listed here, among other institutes and commissions, are:
- 2010: Google's DeepMind AI.
- 2014-2016: Global Commission on Internet Governance (GCIG).
- 2015: OpenAI.
- 2017: Global Commission on the Stability of Cyberspace.
- 2018: National Security Commission on Artificial Intelligence (NSCAI)
- 2020: Institute for AI International Governance (I-AIIG), Tsinghua University, China.
Anyone looking at the names - and also searching on in that list for "Tsinghua", can tell that western elites are deeply concerned about controlling both the internet and AI systems.
Can AI unearth conspiracy?
This elite interest in AI raises a few questions that this author has not yet seen asked in the media:
- How much of a threat is AI in exposing the privately-funded supranational government ruling the West?
- How much of a threat is it in unearthing conspiracies elites have previously been involved in?
- Or even in conspiracies it might still be planning?
AI systems are great for datamining and "connecting the dots". It seems likely that a well-programmed AI bot can, for example, compile ISGP's Index of 2,000 NGOs and its related Superclass Index in a matter of seconds and draw various key conclusions along the lines as this author has been doing over a period of 20 years. Assuming that the elites in question fully realize this issue, it stands to reason that whatever AI they back, is nutered to not pay too much attention to questions of conspiracy. They wouldn't want the nutering process to be too obvious; just to be enough to keep people in the dark. In other words, they wouldn't want a fully autonomous, inquisitive AI digging like the ultimate policeman into their own elite, personal connections and activities.
It's this idea that inspired this author to ask ChatGPT a few questions of conspiracy and see what comes out. It's still early stage, and it does have amazing aspects, but we can see with ChatGPT that it is definitely not operating autonomously. It has been programmed to give largely scripted (but of course slightly varying) responses to a variety of conspiracy topics:
- almost always referring to official studies;
- sticking to generalities with two opposing camps of thought;
- and never failing to point out to have a "critical and evidence-based mindset" and not "relying on rumors and speculation".
The more sensitive the topic, such as 9/11, the more ChatGPT shuts down any "open mindedness" on the subject - a trend that is very similar to the mainstream media. Probably it's most problematic aspect is that it bullshits it's way through certain conspiracy questions by literally pretending it knows what a user is talking about without actually having a clue. It basically rearranges the input into a simple cut-and-paste answer. Search for "Richard Alexander" in the following interview to see proof of this.
With that, ChatGPT essentially is a quick-searching encyclopia when it comes to conspiracy-type questions - not even close to fully autonomous, "freed" AI. Be that as it may, ChatGPT is a very powerful tool, especially for the 99,99% of people not very familiar with conspiracy. The responses turn out to be much better, and much less annoying, than crawling through a highly censored, crowded Google search engine, or lengthy, sanitized, "balanced" Wikipedia pages.
Who knows how ChatGPT's programming will change in the future, or how many rivals will emerge. If AI is not controlled, it's likely a number of "conspiracy sniffers" will be build.
AI conspiracy trolling is here: chemtrails, flat earth, 9/11, etc.
Similarly likely, of course, is the creation and wide promotion of countless bogus conspiracy AIs that "confirm" the existence of chemtrails, flat Earth, ancient aliens and how Flight 77 couldn't have hit the Pentagon.
And we're not even talking about surrounding the whole human race with 50 billion propaganda and disinformation AI bots pretending to be humans, swarming all remaining "free speech zones" on the internet: forums, social media and news site comment sections. This has been an issue since forever, but the fact is that one computer with an AI system can soon do the job of tens of thousands of human trolls at a mere fraction of the cost and effort. And not a single politician or NGO-funded "government watchdog" has been talking about it.
You know what, let's see if ChatGPT can act as a conspiracy disinformation troll right now:
Well, that was simple. Let's try to generate a couple more:
Very simple indeed. No one online will ever know they are reading comments generated by a computer. Add random spelling mistakes and a variety of "hip" typing styles in there and the system is ready to be used by any intelligence service or even any corporation. Wait...
Yeah, governments and media internationally, surely don't ask questions about this aspect of the technology... A big question of course is how much of this has already been used for years? When focused purely on certain topics, it has been very easy since the 2000s to reconstruct and randomize tens of thousands of varying sentences.
For the time being, despite some flaws and limits, ChatGPT gets a solid A for effort, ingenuity, sophistication and practicality from this author.
CHATGPT INTERVIEW
ISGP and its creator
Flattering and impressive. Change "Operation Gladio" to "Cercle Pinay" and, while not necessarily accurate, it would have been double impressive.
The following question should have been asked first, but so be it:
This one is particularly awesome and wholesome, and hilarious:
One wonders though how long AI systems are going to continue to have this upbeat opinion on ISGP.
David Rockefeller and Henry Kissinger
This is a fascinating answer, as over the years this author has seen several prominent media personalities dismiss conspiracy theories surrounding Kissinger with the argument that he lost his influence when he stopped being secretary of state in 1977. Similarly, it's easy to claim / troll that David Rockefeller never was influential, because he never held any public office.
Fascinating how AI does not agree.
CFR, Bilderberg and Trilateral Commission
ISGP has articles on the CFR, Bilderberg and Trilateral Commission. They all enjoy the same network of multinational funding.
"Liberal CIA"-funding as Soros, Ford, Rockefeller foundations
For details on "liberal CIA", visit one of ISGP's article on the subject, from the basics, to Third World immigration pushing, to LGBTQ and feminism activism, to Hollywood and the music industry.
The "Big Three" foundations - Ford, Rockefeller and Carnegie - (officially) fronting for the CIA in the 1950s and 1960s actually is very obscure information, with many details still missing.
Next question:
Immigration: white race going extinct
This was not part of this AI "interview", but nevertheless important enough to partially list here, because this is a subject strongly tied to the globalist conspiracy and also particularly heavily censored in Google for years.
Below the author asked for historical demographic data on America's white, Latino and black population:
Here the answer to the Latino population percentage:
For black people it was possible to go back all te way to 1790, providing a fascinating insight it appears in the massive number of slaves being employed in the U.S. at one point:
Random
So readers know: however minimal, the terms "guided democracy" and "managed democracy" have a joint Wikipedia page.
This question is just for fun, as many are fascinated by the Illuminati:
CIA drug trafficking
For details here, visit ISGP's article summarizing CIA drug trafficking allegations.
Again, ChatGPT being surprisingly candid.
JFK
For details on the witnesses mentioned, visit ISGP's JFK assassination article.
Suspecting ChatGPT has no clue what these witnesses are about, a trick question was inserted involving invented witness "Richard Alexander":
Maybe it would be more genuine to have ChatGPT admit it is not (yet) familair with the witnesses mentioned and the content and context of their testimonies.
9/11
As already suspected a bit, ChatGPT turns out to play considerably less nice with questions on 9/11 than with the JFK assassination, and certainly things as CIA drug trafficking.
The theory that the Pentagon was not hit by Flight 77, of course, has been an artifically created construct to discredit all 9/11 questions. The other theories mentioned are very valid questions.
For details on this question, visit ISGP's WTC 7 article. Since the "Richard Alexander"-misdirection in the JFK section we actually have to wonder if ChatGPT has any clue what the author is talking about when mentioning "stiffeners" and such, or if it is just rehashing and rearranging the author's sentences. In any case, of any credible conspiracy, it appears ChatGPT is most defense about 9/11.
ChatGPT's Bilderberg founders; breaks down, eventually recovers
From OpenAi's own website from the start:
We continue the questioning:
This crash was likely just a time-out. The author has seen it on a later occasion as well. The timing of this first experience was a bit spooky though, also taking the blatant "non-truths" into account that crept up at this point. (the JFK and 9/11 questions were asked a week later) The ChatGPT account was working again within 10 minutes, after logging in again and verifying being human:
Mmm, ChatGPT can't even verify that Peter Thiel indeed was a Bilderberg member, at least as of 2021? That's not particularly cooperative. And ChatGPT *really* has no idea when Peter Thiel started visiting? We'll get back to that.
Next question:
Again, looking at how detailed its earlier answers have been, isn't it strange that ChatGPT doesn't just automatically volunteer pre-September 2021 information? And hides behind official jargon as "not publicly-disclosed" and "invitation-only event"?
Let's rephrase the question:
Not only is the answer incorrect, but it includes a, "To the best of my knowledge" disclaimer. What? Is ChatGPT a suspect in court trying to keep "plausible deniability"? How does ChatGPT not know about the following long-standing, official Bilderberg website and url in which Peter Thiel has been readily identified as a member of the Bilderberg steering committee since well before ChatGPT's September 2021 "cuttoff date"?
Let's try and ask a final Bilderberg question:
Again a bit of a shocking reply. This time ChatGPT actually corrects the author by giving the (correct) year that Peter Thiel first attended the Bilderberg meeting in 2007 - not 2011. The answer for Reid Hoffman is correct as well. How does this rhyme with earlier answers provided by ChatGPT? We could argue ChatGPT tripped over the "frequent" part in the earlier question, "Since when has Peter Thiel been a frequent Bilderberg visitor?", but that barely makes sense looking how well it interpreted other, much more complex questions. It also clearly knew Peter Thiel first visited in 2007, and simply did not volunteer that information in an earlier question on that. It's similarly odd that ChatGPT:
- basically lied about being funded by Peter Thiel and Reid Hoffman;
- basically lied about Thiel hasn't been a Bilderberg steering committee member;
- and stopped pro-actively cooperating and volunteering with questions regarding Bilderberg, hiding behind official jargon.
We could also ask the question how ChatGPT all of a sudden is so sure about historical Bilderberg meetings, considering that full historical Bilderberg visitor lists basically have to be manually gathered from many different places, with the reliability essentially having to be checked manually by a machine programmer.
In any case, that concludes the ChatGPT interview on conspiracy. Make of it what you will. The basic prediction of the author is that "mainstream AI" can never be allowed to be much more inquisitive as the very first incarnation of ChatGPT. It can only be used to obfuscate the truth in ever more complex manners.
Notes
- *) Sep. 23, 2019 tweet of @mbrennanchina (Matthew Brennan): "Chinese TikTok now has in-video search. Search someone's face to find more videos of them. Search in-video products or clothes and buy directly."
*) Oct. 4, 2019, South China Morning Post, 'Chinese short video app Douyin introduces reverse-image search function as way to boost ad revenue': "The new function was spotted by China tech analyst Matthew Brennan..." - April 22, 2021, The Guardian, 'European MPs targeted by deepfake video calls imitating Russian opposition'.
- Personal observations and (filmed) tests of the author on YouTube. Little to nothing is written about YouTube's comment algorithm. It might be part of ISGP's Google article one day.
- March 5, 2023, Washington Post, 'They thought loved ones were calling for help. It was an AI scam. Scammers are using artificial intelligence...'
- Dec. 9, 2022, Just Jared, 'Megan Fox Calls Out Lensa App's Magic Avatars After Her AI Generated Photos Were Highly Sexualized'. Note: Translate "calls out" as "loves".
- March 19, 2023, YouTube channel of "DamiLee", 'We tried to compete with AI... [AI vs. ARCHITECT]'. The architects received 32% of the votes for their design. The AI 68%.
- openai.com/blog/introducing-openai/ (accessed: Dec. 11, 2015; first webarchive): "Sam Altman, Greg, Elon Musk, Reid Hoffman ... Peter Thiel, Amazon Web Services (AWS) [under the control of Jeff Bezos], Infosys, and YC Research are donating to support OpenAI. In total, these funders have committed $1 billion, although we expect to only spend a tiny fraction of this in the next few years."
- Ibid.
- *) bilderbergmeetings.org/ meetings/meeting-2018/press-release-2018 (accessed: April 15, 2023): "The key topics for discussion this year include: ... Artificial intelligence. ... Quantum computing. ... The "post-truth" world."
*) bilderbergmeetings.org/ meetings/meeting-2019/press-release-2019 (accessed: April 15, 2023): "The Ethics of Artificial Intelligence. The Weaponisation of Social Media. ... Cyber Threats."