Propaganda At The Pace Of AI: New AI Tool Shows The Alarming Ease Of Crafting Convincing Fake News

A developer built a "disinformation machine" using OpenAI's ChatGPT, showing everyone just how cheap it has become to spread propaganda on a massive scale.

What Happened: A developer took two months and spent $400 to create a project named CounterCloud, an AI-powered disinformation machine, reported Wired.

See Also: With iPhone 15 Taking USB-C Route, What Good Is Your Lightning Cable For Now?

While the developer chose to remain anonymous, the disinformation tool targeted US adversaries Russia and China.

Going by the pseudonym Nea Paw, the developer used ChatGPT and other image-generation technologies to develop a sophisticated disinformation campaign complete with text, images, and illustrations.

Paw also posted a YouTube video to explain in detail how the CounterCloud project works, concluding that AI disinformation at scale is now extremely cheap and quick to perform.

"Strong language competencies of large language models are perfectly suited to reading and writing fake news articles," Paw said.

It's not exactly the fault of ChatGPT here – Paw used a combination of existing articles, a gatekeeper module, and fake stories, facts, and historical events to stitch together cohesive, authentic-looking stories complete with fake audio clips of newsreaders reading these fake articles.

Why Is It Important: The rise of artificial intelligence (AI) and its widespread availability for the masses has made it easier than ever before for malicious actors to misuse it to spread propaganda and disinformation.

Paw then instructed ChatGPT to like and repost content that agreed with this disinformation profile or counter those that went against its programming.

The CounterCloud project proved to be extremely efficient, too. Paw says it generated "convincing content 90% of the time, 24 hours a day, seven days a week."

A previous study found that AI-based image generators like Midjourney can be easily deceived into generating racist images, but there are ways for you to protect your images from AI.

Image Source- Shutterstock

Check out more of Benzinga’s Consumer Tech coverage by following this link.

Read Next: Apple Is Cooking Up A Secret Recipe To Make The iPhone 16 Display Brighter And More Efficient