ESET announced upcoming AI protection capabilities designed to safeguard how employees interact with AI tools. Demonstrated at RSAC 2026 and set to launch later this year, the new features will expand visibility in the ESET PROTECT Platform to investigate emerging risks tied to everyday AI usage and agentic AI adoption across an enterprise.
“As companies rely more on AI for productivity and automation, they face growing risks around sensitive data exposure, compliance violations, and misleading outputs,” said Juraj Jánošík, ESET Director of Artificial Intelligence. “Agentic AI is shifting the security battlefield back to the endpoint. ESET has spent over 30 years building leading endpoint protection powered by AI and machine learning, so we’re uniquely positioned to help organisations secure this next wave of AI right where it starts.”
As AI tools become embedded in everyday workflows, many employees are using open cloud chatbots without IT oversight, creating “shadow AI” risks and exposing sensitive data such as internal documents, API keys, secrets, and credentials. ESET addresses this through various technologies that get as close to the source as possible, one of which is a secure browser technology that intercepts AI interactions and analyses both prompts and responses in real time, helping prevent data exposure and detect malicious or misleading content before it impacts users.
In demonstrations at RSAC 2026, the new AI protection feature flagged malicious URLs submitted through chatbot prompts, logging activity at the endpoint and surfacing it in the ESET PROTECT Platform for investigation. The same approach applies to prompt injection attempts, scripts, and sensitive data inputs, enabling organisations to block or monitor activity in accordance with their policies. Security teams will gain visibility into how AI tools are used across their organisation through ESET PROTECT Platform logging, helping them investigate risks and enforce policies more effectively.
As organisations expand their use of agentic AI tools, the attack surface is extending beyond chatbot interactions to include emerging AI supply chain risks. These include compromised AI frameworks and tools, such as trojanised components in widely used libraries like LiteLLM, as well as autonomous agents like OpenClaw that can execute actions on a system with limited oversight. ESET has already been protecting its customers from supply-chain attacks through compromised libraries delivered via standard repositories but is noting a rise in these types of attacks and remains committed to further research and development relating to AI tools.
As part of its broader AI security innovation, ESET launched a free ESET AI Skills Checker at RSAC 2026. Available to non-ESET customers and built on the same technology as ESET’s endpoint security products and ESET LiveGuard, the scanner analyses AI skills for hidden instructions, malicious code, and risky behaviour, using multilayered inspection and cloud-based sandboxing. It is currently available as a built-in feature for existing ESET Endpoint users.
For more than 30 years, ESET has pioneered lightweight, high‑performance endpoint security powered by machine learning and artificial intelligence. These new capabilities extend that foundation by helping organisations defend against today’s rapidly shifting threat landscape, where cybercriminals increasingly harness AI to scale attacks, target employees, and automate sophisticated social engineering.
As the only dedicated cybersecurity member of the Agentic AI Foundation (AAIF), ESET is also working to secure emerging AI agent communication protocols through collaboration with industry leaders like OpenAI, Amazon, Microsoft, and Anthropic. Together, the group is working to establish trusted standards, secure protocol designs, and best practices for AI agent interoperability.
Discussion about this post