CVE-2026-4399
RCE in 1Millionbot Millie Chatbot ≤ 3.6.0
Raw vector
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XSummary
CVE-2026-4399 is a high-severity Command Injection (CWE-77) vulnerability in 1Millionbot Millie Chatbot. Its CVSS base score is 8.7 (High).
Operationally, exploitation aligns with the MITRE ATT&CK technique Command and Scripting Interpreter (T1059); ranked at the 18th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as LLM Application Platforms; in the LLM/Generative AI Risks risk domain.
The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.
Deeper analysis AI-assisted summary
Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.
CVE-2026-4399 is a prompt injection vulnerability affecting the 1millionbot Millie chatbot. It enables users to evade chat restrictions through Boolean prompt injection techniques, where a specially formulated question triggers the model to execute an injected instruction upon receiving an affirmative 'true' response. This flaw causes the chatbot to return prohibited information or responses outside its intended context, bypassing containment mechanisms implemented during LLM model training. The vulnerability is classified under CWE-77 with a CVSS v3.1 base score of 7.5 (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:N) and was published on 2026-03-31.
A malicious remote attacker can exploit this vulnerability without privileges or user interaction over the network with low complexity. Successful exploitation allows the attacker to abuse the service for unintended purposes, execute out-of-context tasks, and leverage 1millionbot's resources or even OpenAI's API key, thereby evading original restrictions on chat behaviors.
The INCIBE-CERT advisory at https://www.incibe.es/en/incibe-cert/notices/aviso/multiple-vulnerabilities-1millionbot-millie-chatbot documents this issue among multiple vulnerabilities in the 1millionbot Millie chatbot.
This vulnerability highlights prompt injection risks in LLM-based chatbots, demonstrating how attackers can manipulate model outputs to perform unauthorized actions using provider resources.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-17357
- 🇪🇸 INCIBE: www.incibe.es
Vulnerability Data
Prompt injection vulnerability in 1millionbot Millie chatbot that occurs when a user manages to evade chat restrictions using Boolean prompt injection techniques (formulating a question in such a way that, upon receiving an affirmative response ('true'), the model executes the…
more
injected instruction), causing it to return prohibited information and information outside its intended context. Successful exploitation of this vulnerability could allow a malicious remote attacker to abuse the service for purposes other than those originally intended, or even execute out-of-context tasks using 1millionbot's resources and/or OpenAI's API key. This allows the attacker to evade the containment mechanisms implemented during LLM model training and obtain responses or chat behaviors that were originally restricted.
- CWE(s)
AI Security AnalysisAI
- AI Category
- LLM Application Platforms
- Risk Domain
- LLM/Generative AI Risks
- OWASP Top 10 for LLMs 2025
- None mapped
- AI-specific weaknesses CR
- CWE-1427 — Classic prompt injection: untrusted user input reaches LLM without neutralization.
Mapped by Cyber Resilience · not in NVD. Poisoning and extraction cases are routed to MITRE ATLAS instead of a synthetic CWE.- Classification Reason
- Matched keywords: llm, openai, prompt injection
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V1.2.3V1.2.5V1.2.8V1.2.9
Mitigating Controls (NIST 800-53 r5) AI
Developer testing and evaluation can discover command-construction flaws before deployment.
Input validation directly stops construction of commands from untrusted data containing special elements.
Secure engineering principles include proper neutralization and safe command construction practices.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Secure SDLC practices directly require input validation and neutralization that prevent command injection.
Runtime monitoring of software and data can detect anomalous command execution resulting from injection.
Identifying recorded vulnerabilities enables remediation of command-injection flaws before exploitation.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Secure coding standards require proper escaping and parameterization of commands, directly eliminating CWE-77.
Security testing in development catches command-injection vulnerabilities before release.
Secure development life cycle mandates input validation and command construction practices that directly prevent command injection.
Application security requirements explicitly call for controls against injection flaws including command injection.
Secure architecture principles reduce the attack surface but do not prescribe the specific neutralization techniques needed.
Environment separation limits the blast radius of an exploited command injection but does not prevent the flaw itself.