Hypervisors can be compromised by side-channel attacks where a malicious attacker can be co-located on the same physical server as a vulnerable workload and the attacker will be able to exfiltrate data from the victim with almost no indicators of compromise (IOC).