AI makes it cheaper to produce malware in many variants
Threat actors have long produced slightly different variants of the same malware for each victim to evade signature-based detection. Researchers at Gen Digital describe how generative AI makes this substantially cheaper, calling the phenomenon promptmorphism. They have tracked a loader-as-a-service campaign using the technique, and Bitdefender describes the same approach in its reporting on APT36.
What this means for your business
Detection resting on known file signatures loses value when every recipient gets their own variant. That shifts the weight onto behavioural detection and onto how quickly the organisation spots and handles an anomaly. For leadership, the point is not that the threat is new, but that the attacker's cost has fallen, so volume will rise.
Berigo recommends
- Verify that your endpoint protection actually performs behavioural detection, not just signature lookups.
- Measure your own time from incident to detection and set a concrete improvement target.
- Limit what can execute on clients through application control so unknown binaries do not launch.
- Include AI-assisted attack development in the risk assessment the next time you refresh your threat picture.
Source
Security that is understood, governed and works.
Let us help you turn security into an advantage, not a cost. Get in touch for a no-obligation conversation about where your organisation stands and what to prioritise first.
Get in touch