Researchers develop self-replicating AI worm that can infiltrate, steal data

Researchers have created a computer "worm" that uses generative AI to spread from one computer machine to another. 

If it hasn't already, this is a warning that the technology may soon be exploited to create malicious malware, Independent reported.

This self-replicating AI worm "Morris II" can infect people's emails to transmit malware and steal data.

It was constructed by a multinational team from the United States and Israel to demonstrate the hazards connected with generative artificial intelligence (GenAI) and is named after the first computer worm made in 1988.

The worm is intended to attack AI-powered apps that employ popular technologies such as OpenAI's ChatGPT and Google's Gemini. It has already been used against GenAI-powered email assistants to steal personal information and initiate spam operations.

The researchers cautioned that the worm constituted a new species of "zero-click malware," meaning that the user does not need to click on anything to initiate or transmit the destructive behaviour. Instead, the generative AI tool does the task automatically.

Cornell Tech researcher Ben Nassi said, "It means that now you can conduct or perform a new kind of cyberattack that hasn't been seen before."

The findings were published in a report titled "ComPromptMized: Unleashing zero-click worms that target GenAI-powered apps".