A chatbot is a software interface that uses natural language processing to carry on text or voice interactions with users. In enterprise settings, chatbots can handle first-contact triage, gather information, answer routine questions, or support transactions when they are connected to trusted systems and governed carefully.
Expanded Definition
A chatbot is more than a conversational interface when it is wired into enterprise systems. In NHI security, the key question is not whether the chatbot can respond in natural language, but whether it can initiate actions, retrieve data, or pass requests into authenticated workflows on behalf of a user or an agent. That makes it an access boundary, not just a user experience layer.
Definitions vary across vendors when a chatbot is promoted into an AI agent, but the security distinction is practical: a chatbot that only answers FAQs has limited risk, while one that can create tickets, query records, or trigger payments must be governed as an identity-bearing application. For that reason, controls around authentication, secret handling, approval boundaries, and logging should be evaluated with the same rigor used for other NHIs, especially when the bot uses API keys or service accounts. The NIST Cybersecurity Framework 2.0 is useful here because it frames the chatbot as part of a broader risk-managed system rather than a standalone interface.
The most common misapplication is treating a transactional chatbot as a harmless front end, which occurs when teams expose backend credentials or privileged workflows without explicit authorization checks.
Examples and Use Cases
Implementing chatbots rigorously often introduces a latency and governance tradeoff, requiring organisations to weigh faster user assistance against tighter approval gates and more restrictive tool access.
- A service desk chatbot answers routine password and access questions, but escalation to account changes requires a human approver and an auditable workflow.
- A customer support bot retrieves order status through an API key stored in a managed secret store, rather than embedding credentials in code or configuration.
- An internal IT chatbot opens tickets, checks inventory, and routes requests, but cannot directly modify production systems unless a scoped service account is explicitly authorized.
- A hiring assistant chatbot collects applicant information, but it must not expose resume data or auto-advance candidates without role-based checks and traceable logs, as seen in incidents such as the McDonald’s McHire AI Chatbot Default Credentials case.
- A marketing chatbot embedded in an authenticated portal can personalize responses, but only after session validation and least-privilege access to customer profiles.
Standards guidance for connected systems comes from NIST Cybersecurity Framework 2.0, while NHI governance patterns become clearer in NHIMG research on the Meta AI Instagram Account Takeover, where support automation intersected with account compromise.
Why It Matters in NHI Security
Chatbots matter because they often sit at the point where users, secrets, and privileged actions converge. If a chatbot can authenticate to systems, then its credentials, session tokens, and tool permissions become part of the enterprise NHI attack surface. That is where weak governance turns a simple support feature into a path for data exposure, account takeover, or unauthorized transactions.
NHI Management Group research shows that 97% of NHIs carry excessive privileges and 80% of identity breaches involved compromised non-human identities such as service accounts and API keys. Those numbers are especially relevant when a chatbot is granted broad access for convenience. Poorly scoped tool access, default credentials, and misconfigured integrations also create conditions for rapid abuse, as illustrated by the Schneider Electric credentials breach and the OmniGPT Breach – 34M Conversations Exposed.
Organisations typically encounter chatbot risk only after a support escalation, credential leak, or unauthorized action reveals that the bot had more authority than intended, at which point chatbot governance becomes operationally unavoidable to address.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
OWASP Agentic AI Top 10 and OWASP Non-Human Identity Top 10 address the attack and risk surface, while NIST CSF 2.0, NIST Zero Trust (SP 800-207) and NIST AI RMF set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| OWASP Agentic AI Top 10 | AGENT-04 | Chatbots that act on tools and data are governed as agentic interfaces, not simple UIs. |
| OWASP Non-Human Identity Top 10 | NHI-02 | Chatbots often rely on secrets and service accounts, making secret handling central to risk. |
| NIST CSF 2.0 | PR.AC-4 | Access permissions for chatbot-backed workflows align with least-privilege access control. |
| NIST Zero Trust (SP 800-207) | SC-3 | Zero Trust requires each chatbot request and backend call to be verified, not implicitly trusted. |
| NIST AI RMF | Chatbots introduce AI risk from misuse, overreach, and unsafe outputs that must be managed. |
Authenticate every chatbot transaction and isolate downstream actions with explicit policy checks.
