Monster Hoaxes In The Age Of AI: Seeing Isn’t Believing Anymore

It’s nearly been 100 years since an ancient legend became a mainstream myth. In 1933, after a spate of alleged sightings made their way into the newspapers, people across the world were introduced to the Loch Ness Monster for the first time. Sure, rumors of something strange lurking in the murky waters of the Scottish Highlands had been circulating in oral accounts for centuries, but this was different. This was no longer local folklore; it was suddenly a global “mystery”.

The rest of this article is behind a paywall. Please sign in or subscribe to access the full content.

But what made this monster story different from any other big fish narrative? For one thing, Loch Ness was not unique for supposedly having something strange living in its depths – there are other lochs (which basically means “lake”) in Scotland with similar stories associated with them. The same is true for bodies of water across the UK and Ireland. So why is Nessie different?

There are lots of answers to this question, including the timing and success of its newspaper coverage, but arguably one of the most important factors was its visibility – not long after Nessie’s debut in the headlines, it was supposedly photographed on two separate occasions.

The first photo, taken by Hugh Gray on November 12, 1933, shows a strange shape at the water’s surface. It’s not exactly clear what you’re supposed to be looking at with this picture. According to Gray, it’s the monster, but most people today agree that Gray likely photographed a dog swimming with a stick in its mouth (look at the blurry image long enough and you’ll see it).

The second photo, however, was much clearer. This one was allegedly taken by Robert Wilson, a respected surgeon, on the morning of April 19, 1934, and is probably the most iconic image of the Loch Ness Monster ever created. The photo, known as the Surgeon’s Photo, demonstrates the creature’s long, serpentine neck looming above the water.

The photo solidified the increasingly popular idea that Nessie was some prehistoric dinosaur-like thing, similar to a plesiosaur, living in the loch. The image was so influential that even now, many people who believe in the Loch Ness Monster have a dinosaur-shaped thing in their minds. But from the start, skeptics believed there was something fishy going on with the photo. They just didn’t attempt to seriously debunk it.

It was only in the 1990s that Christian Spurling, one of the people involved in creating the photograph, made a deathbed confession explaining how it was all an elaborate and entertainingly petty hoax. In short, the mastermind behind the hoax was Marmaduke Wetherell, a disgruntled former Daily Mail employee who had been humiliated by the newspaper after they ridiculed him for a previous attempt at fake Nessie evidence. Essentially, Wetherell sold the Daily Mail the Surgeon photo as a way of getting back at them.

This really is a substantial part of the history of the Loch Ness Monster. When you peel through the types of stories or sightings that have perpetuated the myth, what you find are many hoaxes swimming among earnest yet vague reports that can be understood as simple misidentification of things on the loch.

Despite many people claiming to have seen odd shapes or disturbances in water that they believe have no explanation, there has not been any unambiguous evidence of Nessie ever since. The same can be said for pretty much every modern cryptid, like Bigfoot, as well as supposed alien encounters or alleged evidence of giants.

Due to increasingly sophisticated technology, most people now have powerful cameras in their pockets, and yet we still lack clear evidence of these creatures. Sure, every now and then someone claims to have spotted Nessie, Bigfoot, Chupacabra, Skinwalkers, or ghosts on camera, but it’s always distant, blurry, and of questionable provenance.

So, have we seen the end of the grand monster hoaxes of the mid-20^th century? It might be tempting to think that the rise of mobile phone cameras would make such things more difficult, but maybe technology will save them yet. Perhaps developments in artificial intelligence (AI) technologies could create the next generation of (in)credible hoaxes.

Popes and puffer jackets

When asked how generative AI has changed the landscape of misinformation and disinformation, Lisa Given, Professor of Information Sciences in the School of Global, Urban and Social Studies and the School of Computing Technologies at REMIT University, Australia, explained that speed has become a significant problem.

“I think the main thing is really probably to do with the speed at which information can be created and then disseminated, particularly in conjunction with social media and other kinds of platforms,” Given told IFLScience.

“The Pope in the puffer coat you might have seen as a good early example where you know that really whipped around very quickly. And even though within about 24 hours people knew it was a hoax, it had spread far and wide.”

ⓘ IFLScience is not responsible for content shared from external sites.

In 2023, as the late Pope Francis was hospitalized with a respiratory infection, an image of him wearing a large white puffer jacket circulated on Twitter and Reddit. The image, which was created by AI, quickly spread across the internet and convinced many people that the head of the Catholic church had significantly modernized his vestments. 

One way to identify that it was a hoax was to focus on the Pope’s hand. Although it was meant to look like it was holding a coffee cup, the AI technology of the time struggled to generate the complex shape, and so it looks a bit weird.

“But as the technology improves, those tells start to disappear,” Given explained. “It becomes more difficult for people to, at a glance, determine what's real and what's not real. And then combined with the speed at which things can be shared online, it's kind of a recipe for the very fast dissemination of information that's absolutely not true.”

In some ways, the rise of AI-generated mis- and disinformation can be understood less as something entirely new and more as an amplification of a much older phenomenon. After all, hoaxes, fake stories, lies, and misunderstandings have played a role throughout human history. However, Given believes these technologies have introduced challenges that are more unique to our time.

“I think it's an amplification, but it's also enabling deeper reach because the accessibility of that technology really puts it into many people's hands more so than in the past.”

Once things are out there, once they've been shared and replicated at scale, it's very difficult to kind of put the genie back in the bottle.

Lisa Given

It is now very easy for people to create fake or misleading content with little skill (no need to wade into a Loch to take a photo of a crafted fake Loch Ness Monster and hope to get an adequately convincing photo). This, coupled with “the democratisation of knowledge and ability to share openly”, as Given added, makes it easier for anyone to fool other people.

“We're also no longer reliant on news media as a trusted source alone. People are telling their own stories. There's alternative media that pops up, and so that combination [of easily accessible AI tools and alternative media] in the ecosystem is now really very powerful in terms of the reach of fake information.”

And once an idea is out in the wilds of the internet, it becomes very difficult to counter its misleading claims.

“It really depends on the topic,” Given said. “People may not care much about the Pope and what he was wearing enough to continue that storyline, but the challenge of course is that once things are out there, once they've been shared and replicated at scale, it's very difficult to kind of put the genie back in the bottle. And so those images will continue to pop up even when they've been debunked in other corners.

“I think that's part of that amplification. You can't kind of wipe the slate clean in the same way that you might have done 100 years ago.”

The problem with AI monsters

In April 2022, something new appeared to crawl out of the realm of nightmares and emerged into the world. The creature, known as Loab, apparently came into existence while Steph Maj Swanson, an artist, was experimenting with a text-to-image AI model. 

Apparently, Swanson asked the AI program to create an image opposite to what she requested. The result was a strange, uncanny-valley composite entity that is disturbing to look at, something that has since been labeled as the first AI cryptid. 

The creation’s strangeness quickly generated interest across the world, where people helped turn it into a kind of new urban legend with claims that it “haunts the internet”.  

ⓘ IFLScience is not responsible for content shared from external sites.

Loab was not the only AI creature to cause fascination. During the same year, comedian Guy Kelly was toying with AI prompts when he asked the program to create an image of “Crungus”, a made-up name. What came out was a series of strange, monstrous creatures generated by the AI. 

Unlike Loab, which appears to defy efforts to be recreated by other people, anyone could use “Crungus” as a prompt and get their own goblin-like monstrosity.

The fact that AI programs were able to consistently and persistently create entities like Crungus led some people to speculate, as with Loab, whether the AI was simply revealing something real. This was likely a joke, but the case of these two AI things demonstrates the potential AI has to create “convincing” horrors, even though it is largely accepted that these manifestations demonstrate the limits of that generation of AI models.

But what if someone created these monsters to deliberately mislead people? This highlights the distinction between mis- and disinformation, and the intent of the creator.

Misinformation is, as Given says, when “the individual shares information that they may not realise is not true”, such as sharing a photo of the Pope in a puffer jacket. “They thought it was real, but they’ve been misled.”

In contrast, disinformation is when content is shared with the deliberate aim of misleading people from the start. “An example of disinformation might be [the case where] images of a potential strike on the Pentagon were circulated online,” Given explained. “In short order, it was shown to be a hoax, but not before that had actually affected the stock market.”

“I think that's where things like the Loch Ness Monster and Bigfoot, or whatever it might be, become relevant. If someone wants to create a fake image and share it with the world in order to gain notoriety, you know, maybe they want book rights, maybe they want to make a movie, maybe they have some reason behind it that goes beyond a few laughs. That’s when we enter the territory of disinformation.”

So, creating something monstrous through generative AI is not necessarily a bad thing; it’s actually quite fun. But it becomes a problem when someone uses that content to mislead other people.

Interpreted like this, the Loch Ness Monster hoaxes created by Gray and Wetherell back in the 1930s can be seen as classic examples of disinformation. And, all things considered, they were pretty innocuous – after all, no one is hurt by the perpetuated belief that the Loch Ness Monster is real. 

But nearly 100 years later, the socio-political context is different, and mis- and disinformation, especially that created by generative AI, can easily become part of wider bad faith efforts to undermine trust in institutions or individuals.

I think we have to be really mindful that any kind of sharing of that misinformation or disinformation these days can be heightened very quickly and can take on a life of its own.

Lisa Given

“It's not just the image itself, it's that ripple effect that affects [...] the broader context around what we know to be true and who we trust,” Given said.

As many studies have now shown, there are correlations between the belief in cryptids, the paranormal, and conspiracy theories. The link makes sense, given that beliefs in ghosts, aliens, or the Loch Ness Monster involve stepping away from accepted standards of scientific evidence and interpretation and rather relying on anecdotal evidence. 

The same is true for a belief in anti-vax conspiracies or ideas of global cover-ups and so on. Cryptozoology itself is a pseudoscience that is effectively an outgrowth of science fiction and storytelling.

“I think we have to be really mindful that any kind of sharing of that misinformation or disinformation these days can be heightened very quickly and can take on a life of its own,” Given concluded.

When it comes to things like cryptids and generative AI, “there are some people who might be entranced by the initial mythological and folkloric stories [they evoke], who then transfer that way of thinking to other parts of their lives. I think that is really the area where we have to be extremely cautious.”