Acceptable Use Policy
To protect the integrity of the Echelon infrastructure and all enterprise tenants residing on our network, every client must adhere strictly to this Acceptable Use Policy (AUP).
Zero Tolerance for Abuse
Violations of this Acceptable Use Policy may result in:
- Immediate, unappealable suspension of all API keys
- Deletion of deployed agents and associated data
- Termination of your Master Services Agreement without refund
- Referral to law enforcement where required by applicable law
1. Prohibited End Uses
You may not use the Echelon Services — including the Client Portal, API, or underlying inference endpoints — for any of the following purposes:
Illegal Activities
Generating content that facilitates, encourages, or provides instructions on how to perform illegal acts under any applicable jurisdiction.
Malware & Cyberattacks
Attempting to generate polymorphic code, exploit payloads, or orchestrate distributed denial-of-service (DDoS) architectures via autonomous agents.
Disinformation Campaigns
Utilizing Echelon agents to generate mass political disinformation, astroturfed social media engagement, or coordinated inauthentic behavior at scale.
Weapons Development
Modeling, simulating, or optimizing kinetic, biological, or chemical weapon systems using any Echelon compute resources.
Non-Consensual Exploitation
Generating non-consensual intimate imagery, deepfakes, or material that constitutes Child Sexual Abuse Material (CSAM). Any detection of CSAM will result in immediate reporting to NCMEC.
2. Network Integrity Abuse
In addition to content-based restrictions, clients are prohibited from engaging in activities that threaten the stability of the Echelon infrastructure:
Rate Limit Bypass
Attempting to bypass, disable, or circumvent any rate-limiting, authentication, or billing mechanisms.
Compute Exhaustion
Deploying recursive autonomous loops without programmatic exit conditions that cause deliberate localized latency spikes.
Model Extraction
Reverse engineering proprietary algorithms or extracting foundation model weights via adversarial prompting or model inversion.
3. High-Risk Industries
Clients seeking to deploy Echelon AI tooling in highly regulated sectors must inform their Account Executive prior to deployment to ensure the application complies with emerging AI legislation.
Industries requiring pre-approval:
This is particularly important for applications that may be classified as “High Risk AI Systems” under the EU AI Act, the US Executive Order on AI, or similar emerging frameworks.
4. Reporting Violations
If you believe an Echelon client or a deployed agent is violating this AUP, please submit an immediate incident report to our Trust & Safety engineering team.
Trust & Safety Operations
Report violations or request clarification on usage policies.