Ridge Security Packages Autonomous AI Pentesting for On-Premises Use

Ridge Security has launched RidgeOne, a self-contained appliance designed to run autonomous penetration tests and red-team exercises without sending sensitive assessment data to an external AI provider.

The system combines the company's RidgeGen security platform with an open-weight model running on NVIDIA DGX Spark. Ridge Security says the architecture also removes recurring model-token costs, a practical consideration for tests that may run for hours and generate large volumes of model activity.

In a company-run test against OWASP Juice Shop, RidgeOne completed an autonomous assessment in roughly 10 hours and reported 32 verified findings, including 19 rated critical or high severity. It also linked individual weaknesses into multistep attack paths. Those results were produced in a controlled target environment and should not be treated as an independent comparison with human-led testing.

The product reflects a wider shift in AI security tooling toward local execution. For enterprises, the central questions will be whether the appliance can reproduce findings, control potentially disruptive test actions and provide enough evidence for engineers to validate each result before remediation.