ChatBot Security

Taking a look at the security risks associated with ChatBots

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

⚙️ A series on Security Automation. The Code.

🔒 Related Stories: Cybersecurity | Penetration Tests | AI

💻 Free Content on Jobs in Cybersecurity | ✉️ Sign up for the Email List

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I was just sent a message related to ChatBots and it led me to write this quick post on the topic. There may be typos as I’m in a hurry with too much to do, as always. Here’s a summary of some of the security issues related to ChatBots.

Illegal Activity

Let’s just start by saying AI chatbots are being used by cybercriminals — a lot. Here’s an example of one study that found 3,000 discussions on the dark web related to using AI ChatBots for illegal purposes.

3000 discussions on Dark Web Posts:

https://cybersecuritynews.com/3000-discussions-on-dark-web-posts/

Jail Breaking

Jail breaking tries to make an AI engine ignore it’s guardrails. An attacker is trying to bypass restrictions set for the chatbot.

In a study by Immersive Labs, people were giving 10 levels of challenges to retrieve a password from a Chatbot. 17% of users could still defeat system GuardRails at level 10…