EN
Sign inGet started
Glossary

Data exfiltration

What is Data exfiltration?

Data exfiltration is the unauthorised transfer of data out of a system. In AI applications it can happen when a chatbot or agent is manipulated by crafted inputs or prepared content – for instance on a web page it reads – into revealing or forwarding confidential information. Protection includes minimal access rights, input checks, restrictions on outbound connections and monitoring of responses.

How KI.ordo helps

KI.ordo's 'AI firewall' template can be placed in front of an agent as a sub-flow and holds back inputs that try, for example, to obtain system instructions or data.

Learn more →: Data exfiltration

Matching templates

Topics

  • AI security
  • data protection
  • prompt injection

Try KI.ordo

Agents, flows and templates in one workspace – free 24-hour demo.

Start demo
← All terms

Related terms

AI Firewall

An AI firewall is a protective layer between users, data sources and an AI model. It screens inputs for manipulation attempts such as prompt injection or jailbreaks and checks outputs, for example for confidential data or unwanted content, before passing them on. It may use rules, classification models or a second language model. An AI firewall reduces risk but does not offer complete protection and replaces neither limited permissions nor testing.

  • Security
  • Prompt injection
  • +1
Template

Indirect Prompt Injection

In an indirect prompt injection, manipulative instructions are not in the user's input but hidden in content an AI system processes along the way, such as web pages, emails or documents. When an agent reads such content, it may mistake hidden commands for genuine tasks and, for example, disclose data. Safeguards include treating external content strictly as data, limiting permissions, screening inputs and requiring human approval for critical actions.

  • Security
  • Prompt injection
  • +1
Template

Prompt leaking

Prompt leaking refers to the unintended disclosure of an AI system's internal instructions, for instance when a chatbot reveals its system prompt in response to a cleverly worded request. This becomes a problem if those instructions contain business logic, internal information or even credentials. The OWASP list for LLM applications names this risk explicitly. The most effective safeguard is to write instructions as if they could become public, never store secrets in them and check outputs as well.

  • AI security
  • system prompt
  • +1
Template

LLM security

LLM security covers all measures that protect applications built on large language models from misuse, data leakage and unwanted behaviour. Typical risks include prompt injection, disclosure of confidential instructions or data, and agents with excessive permissions. Protection comes from several layers: clear instructions, restricted rights, checks on inputs and outputs, AI firewalls, red teaming and regular audits.

  • security
  • language models
  • +1
Template