CYBERSECURITY GLOSSARY Malware and viruses CYBERSECURITY GLOSSARY Attacks and techniques CYBERSECURITY GLOSSARY Detection and response CYBERSECURITY GLOSSARY Identity and access CYBERSECURITY GLOSSARY Cloud and infrastructure CYBERSECURITY GLOSSARY Compliance and management CYBERSECURITY GLOSSARY Continuity and metrics CYBERSECURITY GLOSSARY Privacy and OSINT CYBERSECURITY GLOSSARY Cybersecurity
← Back to the glossary Cybersecurity glossary · Privacy and OSINT

AI agents and their risks

An AI (artificial intelligence) agent is a system that uses a language model to plan and carry out tasks on its own through connected tools, such as email, files or internal systems. Its main risk is permissions: an agent with broad access can leak data or make changes if it receives hidden malicious instructions (prompt injection) or if no one reviews its actions.

How TecnetOne handles it: Microsoft 365

AI agent security, agentic AI risks, autonomous AI agents, AI agent permissions, agentic AI security

Ready to put this into practice in your operation?

Tell us what you need and a TecnetOne engineer will get back to you.

Talk to an engineer