Microsoft: OpenAI API moonlights as malware HQ

"Microsoft this week detailed a previously unseen backdoor dubbed "SesameOp," which abuses OpenAI's Assistants API as a command-and-control channel to relay instructions between infected systems and the attackers pulling the strings. First spotted in July during a months-long intrusion, the campaign hid in plain sight by blending its network chatter with legitimate AI traffic - an ingenious way to stay invisible to anyone assuming "api.openai.com" meant business as usual."

""Instead of relying on more traditional methods, the threat actor behind this backdoor abuses OpenAI as a C2 channel as a way to stealthily communicate and orchestrate malicious activities within the compromised environment," Microsoft said. "This threat does not represent a vulnerability or misconfiguration, but rather a way to misuse built-in capabilities of the OpenAI Assistants API." Microsoft's analysis shows the implant uses payload compression and layered encryption to hide commands and exfiltrated results;"

SesameOp is a backdoor that abuses the OpenAI Assistants API to operate as a command-and-control channel between infected hosts and attackers. The attack chain begins with a loader that uses .NET AppDomainManager injection to plant a heavily obfuscated DLL. The implant employs Eazfuscator.NET obfuscation, runtime loading, payload compression, and layered encryption to conceal commands and exfiltrated results. The backdoor fetches encrypted commands from the Assistants API, decrypts and executes them locally, then posts execution results back via the same API endpoints. By blending malicious traffic with legitimate api.openai.com calls, the campaign avoids suspicious domains, IPs, and typical C2 indicators.

Read at Theregister