The best way to address segregation of duties issues in an organization with budget constraints is to implement compensating controls, which are alternative controls that reduce or eliminate the risk of errors or fraud due to inadequate segregation of duties. Compensating controls may include independent reviews, reconciliations, approvals, or supervisions. Rotating job duties periodically may reduce the risk of collusion or abuse of privileges, but it may also affect operational efficiency and continuity. Performing an independent audit may detect segregation of duties issues, but it does not prevent them. Hiring temporary staff may increase operational costs and introduce new risks. References: CISA Review Manual (Digital Version), Chapter 2, Section 2.4

The first thing that an IS auditor should do when management responses to an in-person internal control questionnaire indicate a key internal control is no longer effective is to ascertain the existence of other compensating controls. Compensating controls are alternative controls that provide reasonable assurance of achieving the same objective as the original control. The IS auditor should verify whether there are any compensating controls in place that can mitigate the risk of the key control being ineffective, and evaluate their adequacy and effectiveness. The other options are not the first steps, because theyeither require more information about the compensating controls, or they are actions to be taken after identifying and assessing the compensating controls. References: CISA Review Manual (Digital Version)1, Chapter 2, Section 2.2.3

The greatest concern to an IS auditor conducting an audit of an organization that recently experienced a ransomware attack is that backups were only performed within the local network. This means that the backups could have been encrypted or deleted by the ransomware, making it impossible to restore the data and systems without paying the ransom or losing the data. Backups are a critical part of the recovery process from a ransomware attack, and they should be performed frequently, securely, and off-site or in the cloud to ensure their availability and integrity.
The other options are not as concerning as option C, although they may also indicate some security weaknesses. Antivirus software was unable to prevent the attack even though it was properly updated, but this is not surprising given that ransomware variants are constantly evolving and antivirus software may not be able to detect them all. The most recent security patches were not tested prior to implementation, but this is a trade-off between security and availability that may be justified depending on the severity and urgency of the patches. Employees were not trained on cybersecurity policies and procedures, but this is a preventive measure that may not have prevented the attack if it was initiated by other means such as phishing or exploiting vulnerabilities.
References:
* 10: Infrastructure-as-a-Service Security Responsibilities - CloudTweaks
* 5: 3 steps to prevent and recover from ransomware | Microsoft Security Blog
* 7: How to Recover From a Ransomware Attack - eSecurityPlanet

The best way to prevent accepting bad data from a third-party service provider is to implement business rules to reject invalid data. Business rules are logical expressions that define the business requirements and constraints for specific data elements. They can be used to validate, transform, or filter incoming data from external sources, ensuring that only high-quality data is accepted into the enterprise data warehouse. Business rules can also help to identify and resolve data quality issues, such as missing values, duplicates, outliers, or inconsistencies.