Check Point has issued a warning about Russian hackers attempting to breach the AI model ChatGPT. ChatGPT is an OpenAI language model that is being used for a variety of purposes by a growing number of businesses and organizations.
“At Check Point Research, we can see the Russians trying to break through the geo-regional restrictions put in place around ChatGPT,” said Pete Nicoletti, field chief information security officer for Check Point.
According to Check Point, the hackers are attempting to access sensitive information through a variety of methods, including phishing and exploiting software vulnerabilities. According to reports, the hackers are specifically targeting ChatGPT due to its growing popularity and the large amounts of valuable data it processes.
Nicoletti mentioned there’s an attempted exploit called do anything now(DAN), which would use the chat prompt to manipulate ChatGPT into producing text that evades the guardrails put in place by OpenAI to prevent it from producing certain text, such as hate speech. He also stated that he is unsure if any zero-day exploits for ChatGPT have yet to be discovered.
Check Point advises ChatGPT users to take security precautions such as regularly updating software, implementing multi-factor authentication, and training employees to recognize and avoid phishing attempts. The company also advises businesses to limit the amount of sensitive information they process through ChatGPT and to use encryption to protect the data.
The sources for this piece include an article in ZDNET.