AI Misinformation Complicates Search for Missing Child in Australian Outback

As the search for four-year-old Gus continues in the remote South Australian outback, a disturbing trend has emerged online: artificially generated images of the missing boy are circulating on social media, creating additional distress for his family and potentially hampering search efforts.

For two weeks since Gus was reported missing from his family’s homestead approximately 40 kilometers south of Yunta in South Australia’s mid-north, false reports and manipulated photos have been spreading across various platforms. Tech and legal experts have expressed growing concern about the ease with which this deceptive content is being produced and distributed.

Among the most troubling examples is an AI-generated image circulating on Facebook showing a boy with blonde curly hair being held by a man entering a four-wheel drive, accompanied by text questioning if the disappearance is a “kidnapping case.” The same Facebook page has shared over 20 similar posts in the past five days, with some gaining significant traction through thousands of reactions, comments, and shares.

“We’ve got the emotional harm and distress that’s caused by that kind of content being released publicly and being thrown into the public domain when family are emotionally distraught,” said Flinders University law lecturer Joel Lisk. “It might create either false hope or, on the flip side, distress that people are taking advantage of their personal harm and their circumstances for what is effectively clickbait.”

Many Facebook users have condemned the content, with one commenting, “This is sick! Don’t play with people like this, it’s not cool.” When the ABC attempted to contact those responsible for the Facebook page through an email address provided, they received an automatic reply stating the account “does not exist.”

Identifying AI-Generated Images

While generative AI technology is rapidly evolving, there are still ways to identify artificially created images. RMIT computing expert Michael Cowling explains that these tools “scrape together a whole bunch of information about other images” to create new ones.

Telltale signs of AI manipulation often include inconsistencies in lighting, depth, and shadows. “It historically has had trouble with generating hands or positioning limbs in the right place, or smoothing out differences between backgrounds and foregrounds,” Professor Cowling noted.

Even non-experts can often detect what Cowling describes as the “uncanny valley” effect – when an image feels somehow off, even if the viewer can’t immediately identify what’s wrong with it.

Motivations Behind Misinformation

The reasons why someone would deliberately spread falsehoods about a missing child range from malicious intent to financial gain.

“Unfortunately, people can be horrible and they’re doing it to take advantage of a situation and to see how much traction they get with this content,” Dr. Lisk explained. He also pointed out that monetary incentives often drive such behavior: “These horrible websites that exist are normally covered in advertisements. If content creators can develop pages that have high followings and high reach, there is potential there for them to use those platforms to develop revenue through ads or sponsored posts.”

Legal Responses to Digital Misinformation

While some consumer protections already exist that could potentially apply to AI-generated content, experts suggest there’s room for strengthening laws. Dr. Lisk believes lawmakers could “create a law that perhaps prohibits the use of generative AI content in connection with active police investigations.”

However, enforcement remains challenging. “We can serve take-down notices, we can serve cease-and-desist letters, but every time someone posts this content, it spreads like wildfire and you end up with hundreds of different versions of the same misinformation,” Dr. Lisk said.

Professor Cowling emphasized the need for updated legislation: “Is it important that we race to catch up and update our laws on slander and hate speech and misinformation to meet this new reality? Yes, I think we should probably try and do that as quickly as we can because ChatGPT and generative AI is changing the world very quickly.”

Protecting Ourselves from Misinformation

While technical solutions like watermarking AI-generated images could help, Professor Cowling suggests that source verification remains the most reliable method for determining authenticity.

“Understanding the source that something came from — when it was shared, who it was shared by — I think that principle still applies,” he said.

As social media accelerates the spread of information, both accurate and false, developing critical thinking skills becomes increasingly vital. The case of missing Gus serves as a stark reminder of the real human cost when technology is misused during already traumatic situations.