Artificial intelligence is transforming workflows into all sectors, but its massive deployment still raises serious security challenges. At a time when the great language models (LLM) and autonomous agents are integrated into critical processes, the need to shield their operations becomes prominence. Given this scenario, Crowdstrike takes another step by including its Falcon-MCP and AI Red Team tools in the AWS Marketplace, within the new category of AI agents and tools.
This collaboration with Amazon Web Services seeks to facilitate the safe adoption of generative artificial intelligence technologies (Genai), providing mechanisms to test, validate and protect models of AI within the corporate cloud environment. The joint offer allows organizations to operate the AI on a scale without giving up control over their safety, a key challenge in the current accelerated automation context.
Autonomous yes, but with guarantees
Crowdstrike’s new proposal relies on two pillars: on the one hand, Falcon-MCP, a server that acts as a bridge between Falcon’s telemetry and AI-based applications; On the other, the AI Red Team services, focused on detecting vulnerabilities and configuration errors in systems that integrate models such as LLMs.
“The autonomous AI is transforming the businesses of all sizes and sectors, but only a safe AI can be scaled reliably to offer long -term results,” says Daniel Bernard, Chief Business Officer of Crowdstrike.
Jay McBain, Chief of Canals, adds: “Crowdstrike’s work with AWS establishes a new standard on how the ecosystem can operate the AI safely on a scale.”
Safety embedded in the AI life cycle
The integration with key AWS services such as Amazon Sagemaker or Amazon Bedrock reinforces the security proposal from the design phase to the operation of models in production. The AI Team evaluations are aligned with frames such as the Owasp Top 10 for LLM, a reference guide to mitigate risks such as data leaks or unauthorized code execution.
At the same time, Falcon-MCP allows a Plug-And-Play connection with FALCON data, detection, intelligence of threats, user behavior, facilitating the deployment of automated workflows with defensive abilities incorporated from the origin.
Generative AI demands a new protection mentality
The adoption of AI cannot be separated from the mechanisms that guarantee its ethical and safe use. In an environment increasingly governed by autonomous agents and automated decisions, human supervision remains key, but it is also to have tools that allow climbing without losing control. Crowdstrike and AWS offer a joint response to that need: ensure not only data and infrastructure, but also the intelligent systems that manage them.
