CrowdStrike launches new AI security weapon 'Falcon AIDR'… Real-time blocking of prompt injection

Source: TokenPost Original Title: CrowdStrike Launches New AI Security Weapon ‘Falcon AIDR’…Real-Time Prompt Injection Blocking Original Link:

CrowdStrike(CRWD) has officially launched the new feature ‘Falcon AI Detection and Response·AIDR(’ for its flagship security platform ‘Falcon)Falcon(’, marking a full-scale entry into the enterprise AI security market.

Falcon AIDR is designed to safeguard the new security boundary of AI interactions between prompts and agents. According to CrowdStrike, this product is the only solution that comprehensively protects all layers of AI operations, including prompts, models, agents, data, and infrastructure. In particular, the ‘interaction layer)interaction layer(’, where AI performs inference and makes decisions, is a new attack vector that hackers are targeting, and the system focuses on blocking intelligent intrusion attempts aimed at this layer in real time.

Falcon AIDR’s strength lies in applying the architecture of existing endpoint security technology directly to the AI protection domain. CrowdStrike stated that it can monitor and defend prompts, responses, and agent behaviors in real time throughout the entire process, from development to actual deployment. Threats such as prompt injection)prompt injection(, jailbreak)jailbreak(, and sensitive information exposure are also immediately detected and blocked through AIDR.

Additionally, this system features execution log capabilities that allow in-depth tracking of AI usage, supporting compliance and forensic investigations. Its sophisticated policy control and automatic blocking functions help prevent dangerous AI usage in advance. Moreover, a built-in security guardrail feature for developers is provided, encouraging the safe development of AI applications within enterprises.

Michael Sentonas)Michael Sentonas(, CEO of CrowdStrike, emphasized, “Prompt injection is the frontline security threat in the AI era,” and added, “AIDR blocks this threat and fully protects AI interaction processes, making the entire AI infrastructure safer.” He further stated, “With AIDR integrated into the CrowdStrike platform, AI now can be protected within a single, consistent security model across all operational layers.”

Recently, CrowdStrike announced its ‘Enterprise Graph)Enterprise Graph(’ strategy to respond to the AI era, accelerating efforts to enhance AI-based threat detection capabilities. The launch of AIDR is part of this strategy, demonstrating a proactive response to security demands driven by the increasing adoption of AI. Many companies seeking a balance between AI utility and security see Falcon AIDR as a practical solution.