“To reap the benefits of AI, users must have confidence that the AI will behave as designed, and that the results are safe and secure,” the CSA said in the guide. “However, in addition to security risks, AI systems can be vulnerable to attacks by adversaries, when malicious actors deliberately manipulate or trick an AI system.”
The guidelines do not focus on AI safety or broader issues commonly associated with AI, such as fairness, transparency, or inclusion, nor do they address the cybersecurity risks presented by AI systems.
While some recommended actions may overlap in these areas, the guidelines also do not specifically address the misuse of AI in cyber attacks, such as AI-enabled malware, or threats such as fake information, disinformation, and deepfakes, said the CSA.
Source link