External risk intelligence

EnergizeLab Eilik and EilikO FastAPI AI Services Unauthenticated Model Control Vulnerability

CVE advisorySeverity: CRITICAL (CVSS 9.8)

CVE-2026-105038

The vulnerability exists in FastAPI AI inference and profiling services. Such services are commonly deployed as network-accessible API endpoints or web services to allow interaction with AI models, making them reachable via the internet or internal network in standard deployment patterns for AI infrastructure.

Missing Authentication

Halo Surface Signal: 4 out of 5 — likely to be public-facing.

External exposure likelihood

Horizon Alert

Summary of the vulnerability and why it matters

This critical vulnerability affects EnergizeLab's AI services, allowing unauthenticated remote attackers to alter AI model configurations and switch models on the fly. This could potentially impact AI functionality and performance across the organization. The primary concern is to confirm if these specific AI services are in use and assess potential exposure.

Unauthenticated access to AI model controls.
Affects AI services, impacting core functions.
Confirm relevance and assess exposure risk.

Attack Path

How an attacker could exploit the issue

An attacker can reach the EnergizeLab Eilik and EilikO services over the network without needing any credentials. Once accessed, the attacker can manipulate AI model configurations, potentially leading to unauthorized changes and disruption of the AI's functionality.

No authentication required.
Access AI model configuration.
Risk of model manipulation and disruption.

Live Threat

Current exploitation, exposure, and threat context

The EnergizeLab Eilik and EilikO services could allow an unauthenticated remote attacker to alter AI model configurations. This could lead to the attacker switching the active AI model at runtime, potentially affecting the behavior of the AI services. The advisory does not indicate that personally identifiable information (PII) or other specific data types are directly exposed.

AI model configuration and behavior.
Unauthenticated network access.
Altered service functionality.

Operational Fix

Recommended remediation, mitigation, and detection steps

The EnergizeLab Eilik and EilikO FastAPI AI services are likely managed by the platform or AI infrastructure teams. The immediate priority is to identify all instances of these services, assess their exposure and business criticality, and then confirm the accountable owner before planning any remediation.

Platform or AI infrastructure teams own.
Verify service exposure and criticality.
Plan remediation based on identified risk.

Supplementary metadata

CVSS vector

Validate whether this threat affects your internet-facing exposure.

Halo Threat Intelligence helps prioritize remediation with Halo Surface Signal and H/A/L/O context. Start exposure validation with a free external attack surface trial.

Free EASM Trial See How Signal Works

Frequently asked questions

What is the EnergizeLab Eilik AI service?

EnergizeLab Eilik and EilikO are AI-powered robotic platforms. They include FastAPI-based inference and profiling services designed to manage and run AI models. These components act as the backend engine, enabling the robot to process data, perform tasks, and serve model configurations, which makes them central to the device's operational intelligence.

What does CWE-306 mean for CVE-2026-105038?

CWE-306 identifies a Missing Authentication for Critical Function. In the context of this CVE, it means the software performs sensitive operations—specifically managing and switching AI models—without verifying the identity of the user. Because these administrative controls lack a password or login check, any remote actor can interact with them as if they were a trusted administrator.

How does an attacker trigger this vulnerability?

An attacker triggers the bug by sending network requests to the unauthenticated AI administration endpoints on the Eilik or EilikO services. No special preconditions, such as existing user sessions or system access, are required. Simply reaching the service over the network is sufficient; localized actions or physical interaction with the robot are not required to exploit this flaw.

Why should I care if my Eilik services are reachable?

Halo Surface Signal indicates that these FastAPI services are commonly deployed as network-accessible endpoints. If your instance is reachable via the internet or an untrusted internal network, anyone can remotely modify your AI model configuration. This allows an attacker to switch the active model at runtime, effectively hijacking the robot's logic and behavior without your knowledge.

What steps should I take if I use Eilik technology?

First, locate all running instances of Eilik or EilikO services within your environment. Verify whether these services are accessible from your internal network or the public internet. Once you have identified the accountable team, such as your AI infrastructure or platform administrators, restrict network access to these endpoints immediately while you wait for official guidance.

References

Written by Nicholas Merritt

Nicholas Merritt is VP of Security Solutions at Halo Security. With more than 25 years across application security, vulnerability prioritization, and offensive security, he helps teams translate threat data into practical exposure validation and remediation focus. He authors Halo Threat Intelligence CVE advisories using Halo Surface Signal and H/A/L/O context to prioritize internet-facing risk.