Pretty confident it's A here. ISO 27001 considers the information security policy itself a preventive control, not just technical stuff. D is more about detective controls, so that's a common trap in these types of questions. Disagree?
Not a blocker if they're only minor nonconformities, so B is fine. ISO/IEC 27001 lets the audit team recommend certification as long as there are no majors. Had similar on practice sets and that's how it's handled, unless there's a twist in the scenario wording I'm missing.
Had something like this in a mock, B is right. Minor nonconformities don't block the recommendation for ISO/IEC 27001, only majors would. You still need corrective actions but the audit team can recommend anyway. Pretty sure that's standard process, unless guidelines change.
The trick here is that if the external party already has their own strong security controls, option A almost sounds okay. But ISO 27001 still expects explicit contract clauses (B), not just relying on external policies. Pretty sure B is safest unless the question specified minimum expectations.
Actually, I'd pick A here. Weak AI covers systems that are limited to specific tasks like recommendations or data analysis, which is what Skyver wants. Machine learning (B) is a method, but the question asks for the type of AI, not the technique. I might be missing something but that's how I see it.
I don’t think it’s A here. Infrastructure failures usually mean actual hardware breaking or going offline, like if a server dies physically. But the question says it’s about overload/saturation from usage, which points to technical failures (B). Compromise of functions (C) would be more about loss of core business ability or maybe security incidents. Open to counterarguments but B fits best, I think.
Pretty sure it's B. The issue is about system overload due to growth, which hits on technical limitations, not really hardware or total function compromise. Always felt ISO sees this as a technical failure but correct me if I missed something.