Published on February 14, 2024, 5:26 pm

OpenAI and Microsoft have joined forces to take down five state-affiliated threat actors who were using AI services for malicious cyber activities. These actors, originating from China, Iran, North Korea, and Russia, utilized OpenAI models for various tasks including company research, cybersecurity tools analysis, technical article translation, code debugging, and the creation of phishing campaigns.

The actors went as far as attempting to understand publicly available vulnerabilities and conducting open-source research on satellite communications protocols and radar imaging technology. One of the threat actors even tried to locate defense experts in the Asia-Pacific region.

Working closely with Microsoft Threat Intelligence, OpenAI successfully shut down all five of these state-affiliated actors. The collaborative effort between the two organizations demonstrates a proactive approach to combating cyber threats.

OpenAI has stated that these activities align with previous Red Team assessments conducted with external cybersecurity professionals. According to these assessments, GPT-4 (one of OpenAI’s AI models) offers limited capabilities for malicious cybersecurity tasks. OpenAI believes that the use of GPT-4 does not go beyond what can already be achieved using publicly available non-AI tools.

However, reports from cybersecurity companies indicate that sophisticated phishing attacks are on the rise due to advancements in AI technology. Tools like ChatGPT make it easier and faster for threat actors to generate ideas and designs for their attacks.

In a recent experiment conducted by OpenAI, they explored whether GPT-4 could assist professionals or students in creating biological weapons more efficiently than traditional internet tools. The results indicated a small difference in efficiency when using GPT-4 compared to other available resources.

These developments highlight the ongoing battle between advancing AI technology and ensuring its responsible use. As AI continues to evolve, organizations like OpenAI and Microsoft play a vital role in monitoring potential threats and taking swift action against malicious actors.

Cybersecurity remains a critical concern globally. While AI brings numerous benefits and advancements in various fields, it is crucial to maintain a cautious approach and continuously improve security measures to stay one step ahead of cyber attackers.


Comments are closed.