Lasso Security

Lasso Security is a cybersecurity platform purpose-built for securing large language models (LLMs) and generative AI applications. It protects the AI lifecycle from risk management to automated red teaming and runtime threat detection. 

The platform operates at the semantic and reasoning layer of AI applications to help protect organizations from data exposure, prompt attacks, and emerging AI vulnerabilities.

How Lasso Security Works

• Integration: Lasso integrates into enterprise workflows through browser plugins for employee usage tracking or CI and API connections for in-house applications.
• Intent Analysis: User prompts and model responses pass through Lasso's security layers, where the platform analyzes patterns and user intent for potential threats.
• Threat Protection: Lasso detects prompt injections and jailbreaks, masks or tokenizes sensitive data, and validates access controls for RAG systems.
• Response Monitoring: Security-trained machine learning layers analyze interactions with sub-50ms latency to stop anomalies before malicious actions or data leaks occur.

How We Rated Lasso Security

We evaluated Lasso Security based on AI discovery, threat detection, DLP, automated red teaming, agent governance, RAG protection, runtime security, and compliance capabilities. Its intent-based detection and low-latency architecture were also considered.