ESET Previews AI Security Features for Chatbot and Agent Protection

ESET previewed new AI security capabilities at RSAC 2026, announced in a press release. The upcoming features are designed to secure chatbot interactions and AI workflows by scanning both prompts and responses to prevent data exposure, malicious links, and compliance violations.

Built into ESET’s browser security technology, the system intercepts AI interactions in real time, detecting harmful content and blocking uploads of confidential data into public AI systems. The company also demonstrated how the feature flags malicious URLs and prompt injection attempts, integrating with the ESET PROTECT Platform for logging and investigation.

ESET’s new protections extend to AI supply chain risks, targeting compromised frameworks and trojanized libraries used in AI tools. The company additionally launched a free ESET AI Skills Checker, which analyzes AI skills for hidden instructions and malicious code using multi-layered inspection and sandboxing.

As a member of the Agentic AI Foundation, ESET is collaborating with organizations such as OpenAI, Amazon, Microsoft, and Anthropic to help develop secure communication standards and interoperability protocols for AI agents.