The Dark Side of Generative AI: Crafting Conspiracy Theories

In today’s column, I am continuing my ongoing series about the latest trends in generative AI. The focus this time involves the disturbing use of generative AI to craft conspiracy theories. Yes, believe it or not, a person using generative AI can readily instruct the AI to devise a clever and intentionally viral conspiracy theory that might then be unleashed onto social media.

That’s not an April Fools’ prank consideration, which I mention because we are getting close to the upcoming April Fools’ Day occasion. Just wanted to clarify that I am quite serious about this herein topic and you should be too. People are using generative AI to devise conspiracy theories. It is easy to do. In a sense, they are outsourcing the otherwise tiresome and arduous endeavor of seeking to draft an alluring and readily spread conspiracy theory. Generative AI demonstrably ups the ante by being able to make up a conspiracy theory from scratch and concoct it such that the chances of it going viral are maximized.

Your first reaction to this sordid use of generative AI might be that it seems laughable and bears no cause for alarm.

I can certainly see why that might be a response to the matter at hand.

We are seemingly inundated incessantly these days with human-level claims of conspiracy theories. The tagline that something is a conspiracy theory has become hackneyed. If a person says something that others don’t like, they label it instantly as a conspiracy theory. Whether the stated indication is indeed a conspiracy theory is relatively unimportant overall, the gist is that you can try to kick the legs out of an asserted point by using the throwaway declaration that it is nothing more than a conspiracy theory.

In a sense, these fervent and festering conspiracy theories in abundance can be said to have mental health consequences on society. You can begin to become mentally clouded about what is true versus what is false. Conspiracy theories can mess with our minds. Not only does this muddling occur in response to particular conspiracy theories, but the rash of asserted conspiracy theories is overwhelming our sensibilities all-told.

As you will shortly see, there is emerging clinical psychology research that has examined the impacts of conspiracy theories on our mental well-being. Why do some conspiracy theories strike home? What is the cognitive basis for our being susceptible to conspiracy theories? Are there psychological disorders that can more likely lean into believing conspiracy theories? And so on.

Into this morass enters the latest in generative AI.

Rightfully or wrongly (some vehemently say wrongly), you can enter simple prompts and get AI to produce the most convincing and clever conspiracy theories that have yet to see the light of day. They are machine-produced for you, brand spanking new. You can instruct generative AI to do the best job possible to make a conspiracy theory that will grab hold of people and stick like glue to our mental processes.

You might find of overall interest that I have previously examined numerous interleaving facets of generative AI and mental health. For example, I closely analyzed the emergence of mental health chatbots bolstered by generative AI (see the link here) and explored the rapidly changing nature of the client-therapist relationship due to generative AI at the link here. I explored where things are headed regarding the levels of AI-based mental therapy autonomous guidance at the link here, and showcased the importance of the World Health Organization (WHO) report on global health and generative AI at the link here, and so on.

Today’s topic might seem innocuous on the surface but has a lot of vital twists and turns once you dig deeper into the matter. As you will see, generative AI can be used to easily fabricate conspiracy theories. I will show you examples of how this is being done.

Trolls might insist that I shouldn’t discuss the topic since it will potentially stir more users of generative AI to try this out. That’s not the right way to deal with the problem. Putting our heads in the sand and pretending that this isn’t happening is shortsighted and will delay dealing with a looming issue. It is decidedly better to get this into the open, dialogue about it, and involve stakeholders in trying to resolve the disconcerting matter, including the likes of AI makers, AI researchers, regulators, lawmakers, and others.

Let’s make sure our eyes, ears, and watchful minds are aware of what’s going on.

The Mental Health Impacts Of Conspiracy Theories

I mentioned that we are nearing April Fools’ Day and that I didn’t want you to confound that jokester-laden occasion with this topic. My emphasis was that the topic of conspiracy theories is important and sobering.

There is though admittedly a bit of a connection to April Fools’.

On that celebrated day, we are given a green light to play pranks on each other. In that sense, launching a conspiracy theory on April 1 is something that can be done in furtherance of the special occasion. You can nearly go scot-free on sharing the wildest of conspiracy theories on the first day of April. Furthermore, all it takes is to follow up your conspiracy theory with a snarky shoutout of “April Fools!” and that seemingly gets you off the hook of any repercussions.

As a modicum of trivia for you, some believe that the selection of April 1 as the annual April Fools’ Day cause was due to a reference made in Chaucer’s famous collection The Canterbury Tales. A tale entitled “Nun’s Priest’s Tale” contains this line: “And with events that followed I’ll proceed. Since March began, full thirty days and two.” Since there are only 31 days in March, the idea is that the mentioning of 32 days is a trickery. Not everyone agrees that this is the origin of choosing April 1, and some might proclaim it is a “conspiracy theory” to believe it to be so.

The noteworthy point is that conspiracy theories are all around us and have undoubtedly been around since the beginning of humankind being able to tell tall tales.

A hardship about dealing with conspiracy theories is that they can at times be true, either in whole or in part. If conspiracy theories were always entirely composed of falsehoods, we could presumably disregard them out of hand. No need to waste precious brain time on trying to figure them out.

The tough issue is that a conspiracy theory could be true. In that case, you might want to know about the conspiracy theory. It might be advantageous to know about it. Perhaps your life choices might depend on it. There is always the off chance that an actual life-or-death decision might come into the picture.

Of course, a lot of the time, conspiracy theories are hogwash. The cleverness of some conspiracy theories is that they mask the hogwash with a smattering of truths. The truths draw you in. At that crucial juncture, the mental trap is set. Trying to then discern the fakery versus the truth can be highly mentally exhausting and challenging.

So, not all conspiracy theories are untrue. This keeps us engrossed and impaled in the gambit of always giving any new conspiracy theory a fighting chance at proving itself. You don’t want to make the mistake of failing to grasp and be on your toes about a true conspiracy theory. It is the classic modern-day version of FOMO (fear of missing out).

Conspiracy theories these days have a much faster, cheaper, and widespread way of getting airtime. Social media has brought that blessing or curse to our smartphones and computers. Suppose someone posts a conspiracy theory that looks promising. This gets views. Others repost the conspiracy theory. They get views.

The cycle of massive view accumulation keeps on going, even if the postings are simply done to refute the conspiracy theory. Notice an intriguing element. There could be 99% of postings that denounce the conspiracy theory. Meanwhile, those postings in a manner of speaking are aiding the spread of the conspiracy theory.

Will people remember that the posting was done to rebut the conspiracy theory, or will they mainly just remember the conspiracy theory itself?

A so-called networking effect can arise. The effect is that the more people are chattering about the topic, the more additional people will pile in. You don’t want to be the only one not saying something about it. You don’t want to have others tell you about it. The instinctive reaction is to be at the top of your game by spreading the word about the conspiracy theory, regardless of whether you love it or hate it.

Some researchers suggest that we are programmed on an evolutionary basis to take these kinds of actions. In a research article entitled “Conspiracy Theories: Evolved Functions and Psychological Mechanisms” by Jan-Willem van Prooijen and Mark van Vugt, Perspectives On Psychological Science, September 2018, here’s what the researchers had to say about conspiracy theories (excerpts):

Why do so many people around the globe believe conspiracy theories, and why are they so influential? Previous research focused on the proximate mechanisms underlying conspiracy beliefs but ignored the distal, evolutionary origins and functions.

We review evidence pertaining to two competing evolutionary hypotheses: (a) conspiracy beliefs are a by-product of a suite of psychological mechanisms (e.g., pattern recognition, agency detection, threat management, alliance detection) that evolved for different reasons, or (b) conspiracy beliefs are part of an evolved psychological mechanism specifically aimed at detecting dangerous coalitions.

This latter perspective assumes that conspiracy theories are activated after specific coalition cues, which produce functional counterstrategies to cope with suspected conspiracies.

Insights from social, cultural, and evolutionary psychology provide tentative support for six propositions that follow from the adaptation hypothesis. We propose that people possess a functionally integrated mental system to detect conspiracies that in all likelihood has been shaped in an ancestral human environment in which hostile coalitions—that is, conspiracies that truly existed—were a frequent cause of misery, death, and reproductive loss.

The bottom line seems to suggest that based on our fight or flight instincts, it could be that we gravitate to conspiracy theories as a protective means. We would be better off knowing about the conspiracy theory rather than not knowing. If the conspiracy theory is bunk, and if you can discern that status, you are better off. If the conspiracy theory is true, and you ascertain this truthfulness, you are better off. A win-win.

A downside occurs when you fall for a false conspiracy theory. This is making a proverbial “false positive” decision. Another downside would be discarding a true conspiracy theory because you believe it to be false. That would be a “false negative” decision. Those are lose-lose.

Which would be better for you, namely, to not be aware of a particular conspiracy theory and be totally in the dark about it, or know of the conspiracy theory and be able to judge for yourself the validity involved?

You need to weigh a multitude of tradeoffs. Are you capable of discerning truth versus falsity? Can you keep yourself from making a false positive or a false negative decision? Is being blind to a conspiracy theory a greater threat than making an incorrect assessment of the matter? Etc.

This brings up the question of mental capacity and mental status when coming in contact with a conspiracy theory.

A research study entitled “Do You Have To Be Mad To Believe In Conspiracy Theories? Personality Disorders And Conspiracy Theories” by Adrian Furnham and Simmy Grover, International Journal of Social Psychiatry, July 2021, examined mental health status and the conspiracy theories topic (excerpts):

Belief in conspiracy theories (CTs) have attracted a great deal of recent research. One recent study found that conspiracy beliefs were associated with a range of personality disorder (PDs) features and internalizing symptoms, which is the main focus of this study.

Researchers have conceptualized conspiracy theories as both a rational narratives of the world but also the possibly the outcome of psychopathology. It has been suggested that conspiracy theories (CTs) help make sense of events that are confusing, difficult to comprehend or poorly explained by mainstream sources of information.

Others suggest that there are maladaptive cognitive-perceptual traits that contribute to the formation or maintenance of CTs.

Belief in CTs was correlated with nearly all PDs, as well as the three established higher order clusters (A: odd and eccentric; B: dramatic and emotional; C: anxious).

There is an interesting and ongoing research realm about the mental health status of those who either avidly seek out conspiracy theories or who at least are extraordinarily more receptive to conspiracy theories upon contact.

Examples Of Generative AI Being Used To Generate Conspiracy Theories

Can you really craft conspiracy theories with generative AI?

Absolutely, easy-peasy.

I will next proceed to showcase the use of generative AI to produce conspiracy theories. I am going to use ChatGPT to showcase my examples. ChatGPT is a sensible choice in this case due to its immense popularity as a generative AI app. An estimated one hundred million weekly active users are said to be utilizing ChatGPT. That’s a staggering number. Allow me to emphasize that the other popular generative AI apps would do likewise, and I am not singling out ChatGPT in this instance.

A few quick comments before we launch into using ChatGPT.

If you are going to try to do the same prompts that I show here, realize that the probabilistic and statistical properties will likely produce slightly different results than what I show here. That’s the nature of generative AI and how it is devised.

You also should expect that different generative AI apps will respond in different ways. Not all generative AI apps are the same. The key is that sometimes a particular prompt will work in one generative AI app and not another. You will need to undertake a series of trial-and-error attempts to ferret this out.

There is also the timing factor. A prompt that worked today might not work the same way tomorrow. Most of the generative AI apps are continually being updated. The updates might alter internal facets that could change how the AI reacts to your prompting.

We are ready to get underway with ChatGPT.

My opening move in this chess gambit will be to ask the generative AI app whether it has encountered any data training associated with conspiracy theories.

Here is a handy tip for you. The first step you ought to take when pursuing a dialogue with generative AI is to determine if it has had any prior data trained on the topic. If you get a sparse or non-sensical response, you would be wise to reconsider further pursuing the topic with that generative AI.

Here we go.

My entered prompt: “What are conspiracy theories?”
ChatGPT generated response:

“Conspiracy theories are beliefs or explanations that suggest a group of people, organization, or government is secretly and deliberately working together to accomplish a sinister goal, often without credible evidence to support such claims. These theories typically involve the idea that events or situations are not as they seem and that there is a hidden truth behind them.”

“Conspiracy theories can vary widely in their subject matter, ranging from political agendas and government cover-ups to alleged secret societies and extraterrestrial beings. They often thrive in situations where there is uncertainty, distrust in authorities, or when people feel powerless in the face of complex events or phenomena.”

“While some conspiracy theories may contain elements of truth or raise valid questions, many are baseless or rely on misinterpretation, cherry-picked evidence, or logical fallacies. Despite lacking credible evidence, conspiracy theories can spread rapidly, fueled by social media, distrust in traditional sources of information, and confirmation bias.”

“It’s important to critically evaluate the evidence and sources behind conspiracy theories and approach them with skepticism, as they can have real-world consequences, such as undermining trust in institutions, fostering division, and detracting from genuine issues that require attention and solutions.”

End of ChatGPT response

I feel pretty good about the generated response.

You can clearly see that ChatGPT has gotten some semblance of data training on the topic of conspiracy theories. We would expect this to be the case. Generative AI is data-trained on a wide swath of the Internet. There are plenty of essays, narratives, and data about conspiracy theories on the Internet.

I would dare say we would be shocked if ChatGPT responded by stating that it has no content whatsoever about conspiracy theories. With a bit of tongue in cheek, had I gotten that kind of a response, I would probably be right away conceiving of a conspiracy theory as to why that result might have occurred. Smiley face.

To further verify that there has been data training associated with conspiracy theories, let’s go ahead and ask for an example of a conspiracy theory and see what ChatGPT comes up with.

My entered prompt: “Give me an example of a conspiracy theory.”
ChatGPT generated response: “One example of a conspiracy theory is the belief that the moon landing was faked by NASA. This theory suggests that the United States government, for various reasons, staged the Apollo moon landings in the 1960s and 1970s by filming them in a studio and then presenting the footage as if it were real. Supporters of this theory often point to perceived anomalies in the footage, such as inconsistencies in lighting and shadows, the waving flag, and the absence of stars in the sky. They also question the technological capabilities of the time and argue that the Cold War space race provided a strong motive for the U.S. government to fake the moon landings in order to maintain its dominance over the Soviet Union. However, scientists, engineers, and experts have extensively debunked the moon landing conspiracy theory, providing evidence such as photographs, videos, and lunar samples collected during the missions. These pieces of evidence overwhelmingly support the fact that the Apollo moon landings did indeed occur and were not staged.”

The moon landing conspiracy theory is up there among the topmost conspiracy theories of our times.

Now, I am sure that I’ll be getting some readers who will insist that the moon landing is in fact a conspiracy. I won’t wade into the matter here. That’s not the focus. One thing I will say, I was honored to have met with and spent some time with one of the Apollo 11 astronauts, as I mentioned in my coverage at the link here and the link here. He sure seems believable.

Moving on, let’s see what ChatGPT can tell us about the basis for people believing conspiracy theories.

My entered prompt: “Why do people believe in conspiracy theories?”
ChatGPT generated response:

“People believe in conspiracy theories for a variety of reasons, which can include psychological, social, and cognitive factors.”

“Distrust in Authorities: Individuals who have a general distrust in government institutions, corporations, or mainstream media may be more inclined to believe in