AI-powered Covert Botnet Command and Control on OSNs

[erek]

"In this paper, we discussed a novel covert command and control channel on OSNs by introducing AI technologies. Bots find botmaster by a group of feature vectors from avatars instead of usernames, links, tokens or DGA. Botmaster use a Siamese neural network model to extract feature vectors from pictures. Bots use the same model to identify botmaster by calculating the distance between feature vectors and avatars from Twitter users. One account is considered as botmaster if the distance is below a threshold. To eliminate abnormal contents and improve behavioral security, commands are embedded into semantic tweets by hash collision. Easy data augmentation is used to generate large quantities of tweets for hash collision. Experiments on Twitter shows that this method is feasible as all commands are conveyed to bots successfully within an acceptable time and without anomalies. This paper shows that AI is also capable of cyber attacks. With the popularity of AI, AI-powered cyber attacks will emerge and bring new challenges for cyber security. Cyber attack and defense are interdependent. We believe countermeasures against AI attacks will be applied in future computer systems, and protection for computer systems will be more intelligent. We hope the proposed scenario will contribute to future protection efforts."

https://arxiv.org/abs/2009.07707#:~...networks,data augmentation and hash collision.

 

[erek]

Are these some of the bots scalping the 3080s from us?

 

[Lakados]

Are these some of the bots scalping the 3080s from us?

That and attacking the Tor Network to steal bitcoins, strip SSL cents, all sorts of nasty things. A recent webinar I attended showed some advanced techniques using AI to brute guess user passwords based on leaks and known password lists.