BlackFog's new ADX Vision 2.0 is designed to tackle that problem at the endpoint, inspecting every AI prompt before it reaches the model, whether it comes from an employee or an AI agent.
The latest release introduces several important capabilities:
Prompt Protection helps detect suspicious or risky content before it is sent. LLM redirection can steer users away from unapproved AI services and towards sanctioned platforms. Prompt Auditing gives organisations greater visibility into how AI is being used, while Token Visibility provides insight into consumption, usage patterns, and potential misuse. The result is a much stronger governance layer around AI adoption.
For Microsoft partners, this is particularly relevant. Customers adopting Microsoft 365 Copilot, Copilot Studio and wider AI services are increasingly asking how they can enable AI without losing visibility or control.
BlackFog can sit alongside those Microsoft conversations as an additional security layer, helping partners move the discussion from “how do we deploy AI?” to “how do we deploy AI securely?”
Hammer has also refined the BlackFog commercial model, making it a more compelling and straightforward attach-sell for Microsoft partners. As AI adoption accelerates, controlling the prompt could become just as important as protecting the endpoint.
Read more about Blackfog