From the course: Managing AI Security Risks with ISO 27001

Unlock the full course today

Join today to access over 24,900 courses taught by industry experts.

Understanding the landscape of AI security

Understanding the landscape of AI security

From the course: Managing AI Security Risks with ISO 27001

Start my 1-month free trial Buy for my team

Understanding the landscape of AI security

“

- Now let's take a look at the AI security landscape. Actually, let's go over the current state of AI security with its common threats and the importance of standardization, like the one we are talking in this course. The current state of AI security is a battlefield of innovation, where defenders and attackers continuously adapt to the evolving landscape. It's a race between enhancing AI's potential for good and mitigating the risks that come with it. Here are currently three major threats to an artificial system. First, data poisoning. As the threat name suggests, this is a form of an attack that targets data that is being used to train the AI models. By injecting malicious or misleading information into the training data set, attackers can manipulate the AI to make incorrect decisions, exhibit biased behavior, or fail at its intended task. Imagine a recommendation system providing awful recommendations to you, or even better, imagine an autonomous vehicle making a decision based on…

Contents