Home › Forums › AWS › AWS Certified AI Practitioner AIF-C01 › Observations on AIF-C01 Timed Mode Set 1 › Reply To: Observations on AIF-C01 Timed Mode Set 1
-
Hello Benjamin,
We are currently addressing the question you mentioned earlier, and updates will be provided as soon as possible.
It appears there was some confusion regarding question 33. Question 33 is focused on Amazon SageMaker inference options and does not involve services like Amazon Q for Business or Amazon Lex.
Furthermore, regarding question 20, while I see your point about DoS (Denial of Service) attacks affecting the availability of the model, model theft indeed compromises the integrity of the AI model.
Model theft has serious consequences for the integrity and overall security of the AI system. Attackers can reverse-engineer a model, analyze the internal decision-making process, and discover vulnerabilities when a model is stolen. This can lead to unauthorized modifications or model misuse, affecting its intended behavior. Furthermore, a stolen model can be repurposed for malicious actions, which ultimately impacts its integrity as it can no longer be trusted to function as designed.
Although DoS affects availability, it is more of a secondary concern in terms of performance or the fundamental integrity of the model. The focus here is on ensuring the model continues to perform reliably and securely without being compromised or stolen.If you need further assistance, please don’t hesitate to message us. Thank you
Regards,
Nikee @ Tutorials Dojo