The correct answer is Contextual risk assessment and decision making.

Contextual risk assessment and decision making is least suitable for full automation because it depends on business priorities, legal and regulatory nuances, and stakeholder risk tolerance that require human judgment. Automated systems can collect data and score risk factors but they cannot reliably make the trade off decisions that align with strategy and compliance obligations.

Contextual risk assessment and decision making can be assisted by automation through aggregated telemetry, vulnerability data, and decision support, but the final decisions and exceptions usually need human oversight and accountability.

Automated vulnerability scanning is not correct because scanning is a routine technical task that is readily automated and it can trigger follow up workflows for remediation.

Telemetry aggregation is not correct because collecting, normalizing, and routing logs and metrics is inherently repeatable and scales well with automation.

Periodic compliance reporting is not correct because evidence collection and report generation for regular audits follow defined templates and can be automated to produce consistent outputs.

Full AWS Practitioner Certification Question