Experts warn 'fun' ChatGPT caricature trend may not be as harmless as you think

A new artificial intelligence trend is rocketing onto social media feeds around the world – and it could pose a risk to your personal data.

Millions of people are prompting ChatGPT to design a caricature of themselves using what it knows about their personality and career.

The result can be fun and whimsical, and sometimes a little uncanny.

READ MORE: Sophisticated 'deepfake' AI is making romance scams more sophisticated than ever

But AI and cyber experts are asking the same question: Is it harmless fun or facilitating a mass privacy crisis?

Digital sociologist and senior lecturer at Australian National University (ANU) Jessamy Perriam told nine.com.au the trend is "fun but eerie at the same time".

"It's more about how much do you want ChatGPT to know about you?" Perriam said.

"You're not really entirely certain whether what you've disclosed to ChatGPT will just be kind of siloed away to just you, or whether it will use it in other scenarios."

Perriam said she was surprised to see the trend take off when there is a cultural shift against the pervasiveness of "AI slop" on social media.

Consumers have pushed back against brands or celebrities using AI in their content.

READ MORE: Sydney couple sold house, quit their jobs after feeling 'trapped' in rat race

A 2025 Neilson survey found 55 per cent of audiences are "uncomfortable" with the use of AI in media.

Perriam urged people to use this critical lens for their own AI use and to be wary about what details they willingly feed ChatGPT.

"Ask yourself, is this something that I want to participate in, is it something that provides me value and enrichment in my life?" she said.

"It calls for this need for people to be really informed about what AI is about, because it's a pretty fuzzy term at the best of times.

"I would be careful about where and who you disclose personal information to online, regardless of whether it's ChatGPT or not."

Trademark and AI lawyer Jessica Eaves Mathews argued in her Substack that the caricature trend was feeding an insatiable big-tech machine.

She warned that users were freely providing the tools to allow institutions to construct a "profile" on them.

"It's easy to see the appeal. It's creative. It's flattering. It feels like self-expression. It's shareable," Eaves Mathews wrote.

"This is why I don't think the real question is whether these AI caricatures are cool or cringe.

"The real question is whether we still have the awareness and self-discipline to resist trends that actually harm us and strengthen big tech's hold on all of us, including our institutions, government and rule of law."

READ MORE: Thousands of Aussies could be risking their jobs with this AI policy blunder

Eaves Mathews said users were giving up intimate details of themselves to a company which was "collecting, processing, and leveraging data at enormous scale".

"This is why it's reckless to treat ChatGPT like a diary, a confessional, or a personality mirror," she said.

"It is not your friend. It is not a therapist. It is not a private sketchbook."

Every ChatGPT caricature requires a certain amount of electricity, too.

According to the MIT Technology Review, generating an image using AI can take the same amount of energy as fully charging a phone.

Perriam said the environmental cost tends to be the forgotten consequence of mass AI trends.

"If everyone's generating cutesy caricatures of themselves, it uses a lot of electricity, a lot of data centre energy, and a lot of water to do all of this," she said.

"It's really environmentally dicey to, a) establish these trends and then, b) have everyone just jump on the bandwagon."

NEVER MISS A STORY: Get your breaking news and exclusive stories first by following us across all platforms.

• Download the 9NEWS App here via Apple and Google Play
• Make 9News your preferred source on Google by ticking this box here
• Sign up to our breaking newsletter here