Researchers at the University of Toronto have created the world’s first adaptive AI-powered computer worm as a proof-of-concept malware in a yet-to-be peer-reviewed study posted in arXiv.
This virus can learn on its own, make decisions, and change its attack strategy in real time to target any device connected to your network. It might sound like something out of the Terminator, but it’s not all bad news.
Related
These 4 antivirus apps are actually worse than malware
Your antivirus should protect you, not make your computer feel more difficult to use.
What’s the difference between this and viruses?
Traditional malware uses fixed scripts to attack certain parts of your computer. When you think of normal malware, you probably think of a virus, which needs you to open an infected file to take control. Worms are a bit different – they can spread across networks on their own and copy themselves, taking control without any input from you.
The autonomous worm from the University of Toronto uses free AI models to study its target, whether it’s a desktop, laptop, or phone. It then picks known weaknesses to attack, using large language AI models like ChatGPT that are available to the public.
It gets even more concerning: this worm costs hackers nothing and takes almost no time to use. Worms that use AI chatbots can adapt, collect information, and attack new targets without any human help. After infecting a device, the worm uses that device’s processing power to run its AI.
How to protect yourself from adaptive malware
While it may sound like a science-fiction threat, the researchers at the University of Toronto have some reassuring news.
“The worm parasitically uses compromised machines to run open-access large language models (LLMs) to sustain its reasoning, or extend its reach for further attacks”.
“It was imperative for us to understand this threat in a controlled, academic setting before bad actors figured it out for themselves,” said Nicolas Papernot, author.
“The reason we are doing this research is to ensure the security of the digital ecosystem we all rely on – to keep people safe. This finding catapults us into a new era of cybersecurity,” Dr Papernot said.
You can take steps to protect yourself
Since these adaptive worms spread by finding weak spots and human mistakes, the best defense is to strengthen your personal digital habits.
Remember that devices like smart cameras, routers, and WiFi-connected appliances are also at risk. Change all default passwords from the manufacturer, and if you can, put your smart home devices on a separate guest Wi-Fi network.
You can also set up stronger access controls. According to new guidelines from a top spy agency, passkeys are now the most secure way to protect your data, and everyone is encouraged to use them. Also, turn on multi-factor authentication on any device that allows it to keep your personal and financial accounts safe.
Finally, don’t ignore software updates. It’s easy to put them off, but updates fix the exact bugs that AI-powered worms look for. Treat operating system and firmware updates as urgent and important tasks.
