Topic starter
10/07/2026 9:08 pm
TL;DR: AI is being adopted broadly in cybersecurity, with 93% of security leaders reporting GenAI use and 91% using it for cyber work, while certificate management teams are increasingly using it to automate issuance, renewal, revocation, and anomaly detection, according to GlobalSign. That efficiency also creates a new governance problem: machine assistance can weaken trust assumptions in PKI if data poisoning, prompt injection, or over-automated approvals are not constrained.
Editorial analysis by NHI Mgmt Group, based on content published by GlobalSign: “Die versteckten Risiken von KI bei der Zertifikatsverwaltung”.
By the numbers:
- In 2024, 93% of security leaders said their companies use GenAI, according to GlobalSign.
- In 2024, 91% of security leaders said they use GenAI specifically for cyber work, according to GlobalSign.
Key questions
Q: What breaks when AI is allowed to influence certificate issuance decisions?
A: The failure is not just automation error.
Q: Why does AI increase the operational risk of poorly managed PKI environments?
A: AI raises risk because it speeds up both attack activity and the scale of digital identities that defenders must manage.
Q: How can security teams tell whether AI-assisted certificate controls are working?
A: Look for whether AI recommendations are traceable to approved data sources, whether every issuance still passes policy checks, and whether anomalous certificate actions are reviewed before trust is extended.
Practitioner guidance
- Separate recommendation from authority Allow AI to suggest anomalies or renewal priorities, but require deterministic policy checks before issuance, revocation or trust-store changes.
- Harden model input paths Restrict the data sources that can influence certificate workflows, and monitor those inputs for poisoning, prompt injection and malformed metadata.
- Audit certificate approval boundaries Map where human approval is still required and where AI outputs are allowed to trigger action, then remove any implicit approval paths.
Bottom line: AI makes certificate management faster, but it also brings the model and its inputs inside the PKI trust boundary.
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
10/07/2026 9:32 pm
AI-assisted certificate management turns model integrity into a PKI control requirement: once a model influences issuance, renewal or revocation, the integrity of the model's training data and prompts becomes part of the trust boundary. That is a governance shift, not just an efficiency gain. The implication is that certificate workflows now need evidence of model control, not merely operational automation.
A question worth separating out:
Q: What happens if a compromised model is allowed to approve certificates?
A: A compromised model can legitimise malicious certificates, expose sensitive PKI details or enable phishing and man-in-the-middle activity under trusted infrastructure. Once that happens, the issue is no longer a model defect. It becomes a direct compromise of certificate trust and identity assurance.
👉 Read our full editorial: AI in certificate management changes PKI risk and governance