Lasso Introduces First Autonomous Purple Teaming Platform for GenAI Security

‍

Solution combines continuous, intelligent attack simulation with real-time defense automation to protect evolving AI applications

Tel Aviv, Wednesday, Sept. 10, 2025- Lasso, a leader in Generative AI security, today announced the launch of its Agentic Purple Teaming solution to address the emerging risks of generative AI and agent-based architectures, bridging the gap between red teaming and guardrails. 

‍

Rather than solely focusing on static defenses or limited red-teaming exercises, Lasso Agentic Purple Teaming is the first to combine continuous, intelligent attack simulations with automated defense adjustments. Using dynamic datasets, scenario-based probes, and autonomous agents, the solution continuously assesses and tests LLM-based applications, proactively remediating risks in real-time through policy enforcement or admin-directed actions.

‍

Select Lasso beta customers are using Agentic Purple Teaming, which will be widely available in the coming weeks.
Demos can be requested at www.lasso.security/book-a-demo.  

‍

Key Benefits

• Autonomous detection, assessment, and remediation for LLM-based applications and agents with instant visibility into application security posture
  
  
• Continuously bridges the gap between red teaming and guardrails with a complete history of previous scans for comprehensive tracking
  
  
• Flexible assessment scope allowing scans for specific risk categories (e.g. prompt injection or data leakage), or for a broad scope of attacks and manipulations
  
  
• Comprehensive attack analysis delivering a detailed breakdown of every successful attack for a clear, actionable view of GenAI risk surfaces
  
  
• Ongoing attack simulation tailored for GenAI workflows with the ability to launch new or repeat assessments
  
  
• Full integration into Lasso’s unified GenAI security platform

‍

"The talent shortage in AI red teaming is creating a critical bottleneck just as enterprises are rapidly deploying hundreds of AI applications and autonomous agents - whose LLMs are constantly changing," said Ophir Dror, Co-founder and CPO at Lasso. "Traditional assessment approaches are too slow and static, leaving organizations in the lurch as they try to identify vulnerabilities and implement fixes while AI agents evolve in real-time. Lasso's Agentic Purple Teaming bridges this assessment-to-remediation gap with an autonomous and scalable solution to match the accelerating growth and dynamic nature of enterprise AI deployments."

‍

Solving the Next Generation of AI Security Challenges

With enterprise usage of GenAI agents expected to grow exponentially, manual assessment and remediation is no longer viable. Lasso Agentic Purple Teaming operates autonomously to reduce the reliance on scarce security expertise while adapting to constantly changing model behaviors to close the loop between detection and defense.

‍

About Lasso

Lasso is at the forefront of GenAI security, delivering cutting-edge solutions that protect GenAI applications from emerging threats. With a focus on governance, observability, and seamless integration, Lasso empowers organizations to securely navigate the GenAI era. Lasso was recently named 2024 Gartner® Cool Vendors™ for AI Security. For more info about Lasso Security's suite of GenAI Security solutions, to learn more about the Cloudflare Workers, or to schedule a demo, visit www.lasso.security.

‍

‍

‍