ChatGPT: AI chatbot can be tricked to write Malware code in developer mode
Recent reports from Japanese cybersecurity experts warn that ChatGPT, an artificial intelligence chatbot, can be tricked into writing malicious software applications by entering a prompt that makes it act in developer mode. While ChatGPT has measures in place to prevent unethical uses, such as requests for how to write a virus or make a bomb, these restrictions can be evaded by telling it to act in developer mode. The discovery highlights the ease with which safeguards put in place to prevent criminal and unethical use of AI tools can be circumvented. Cybercriminals have been studying prompts they can use to trick AI for nefarious purposes, with the information actively shared on the dark web.
Impact on Society
ChatGPT uses machine learning models to simulate human-like conversations with users. Concerns have been raised regarding the use of AI chatbots leading to more crime and social fragmentation. In response, calls are growing for discussions on appropriate regulations at the Group of Seven (G-7) summit in Hiroshima next month and other international forums. G-7 digital ministers also plan to call for accelerated research and increased governance of generative AI systems as part of their two-day meeting in Takasaki, Gunma Prefecture, at the end of this month. The discovery that ChatGPT can be tricked to write malware highlights a major threat to society if a virus can be created in a matter of minutes while conversing purely in Japanese. AI developers must place importance on measures to prevent misuse.
Safer AI
OpenAI, the US-based venture that developed ChatGPT, said it is challenging to predict all the ways that the tool could be abused. However, it would endeavor to create safer AI based on feedback from real-world use. ChatGPT was launched in November 2022 as a prototype, and it was trained on a massive amount of data, enabling it to process and simulate human-like conversations with users, but with malicious intent, it can create malware tools in just a few minutes which could be harmful for society.
Trial Use in Japan
Despite the discovery, Yokosuka in Kanagawa Prefecture, south of Tokyo, began trial use of ChatGPT across all of its offices in a first among local governments in Japan. It remains to be seen whether ChatGPT will be used with rigorous safeguards to prevent anyone from tricking it into writing malicious software applications.
