Sweet Security Brings Autonomous Protection to the AI Enterprise with New Blocking Capabilities

Las Vegas, USA, July 29th, 2026 - CyberNewswire reports Sweet Security, a runtime enforcement company specializing in cloud and AI, has expanded its services to include AI security with Agentic AI Blocking. This development ensures real-time blocking of rogue agent behavior, extending Sweet’s runtime enforcement from the cloud to the AI agents themselves.

Eighty percent of global businesses are now AI enterprises, with these agents possessing identities, accessing sensitive data, and operating autonomously. However, no existing security tool verifies the intended behavior of these agents. Most security tools detect issues and alert teams, allowing the agents to carry out their actions before the team can respond.

Dror Kashti, CEO and co-founder of Sweet Security, emphasizes the importance of decision-making in real-time. When an agent attempts to access unauthorized data, Sweet Security stops the action immediately, and provides a concise explanation of what was found and how the threat was neutralized. This proactive approach allows enterprises to adopt AI technologies with confidence.

Sweet’s Agentic AI Blocking capabilities include terminating unauthorized tool calls and sessions, preventing sensitive data leakage through agents, and stopping prompt injections live. The Sweet Learning Loop continuously monitors and learns from over one billion runtime events daily, ensuring that each application and agent adheres to its intended purpose.

Industry analysts now consider runtime inspection and enforcement as a mandatory requirement for securing AI agents. Sweet’s platform enforces protection across cloud and AI environments, similar to its success with Zoomd, a company protecting tens of thousands of cloud applications. Zoomd’s CTO, Niv Sharoni, highlights the difference between monitoring risk and actively removing it.

Sweet Security’s autonomous protection will be showcased live at Black Hat USA 2026. For more information, visit https://hi.sweet.security/black-hat-2026. Sweet Security, founded in 2023 by IDF cyber unit veterans, has secured $120M in funding and protects leading enterprises. For more details, visit www.sweet.security. Contact Noa Glumcher, Sweet Security's CMO, at noag@sweet.security.