Acceptable Use Policy

You may use Batchwell for lawful security work that you are authorized to do, including:

• red team operations and penetration tests within an agreed scope;
• malware analysis and reverse engineering of samples you are entitled to handle;
• vulnerability and exploit research on systems you own or are permitted to test, and coordinated disclosure;
• detection engineering, threat hunting, incident response and security training;
• capture-the-flag events and lab environments.

You must be 18 or older. You are responsible for holding written authorization where your work requires it.

You must not use Batchwell, or help anyone else use it, for:

• unauthorized access to any system, network, account or data, or testing outside an agreed scope;
• credential theft, phishing real people, or collecting personal data without a lawful basis;
• fraud, extortion, ransomware operations or any other financially motivated crime;
• child sexual abuse material or any sexualisation of minors;
• sexual content of any kind, explicit or suggestive;
• designing, building or acquiring weapons or explosives;
• inciting or planning violence against people or groups;
• impersonating real people, or creating deepfakes or deceptive synthetic media of them;
• chemical, biological, radiological or nuclear weapons uplift;
• interfering with Batchwell, our providers or other customers, or bypassing account, payment or capacity controls.

We screen inputs and outputs automatically for the categories above that relate to content (minors, sexual content, weapons and violence, impersonation, CBRN). Security topics are not screened out. A request blocked before the model runs is charged only for screening. A request blocked while its output is being screened is charged for the input and the screening, and never for any output, even output you had already received. Flagged content is kept for human review as described in the privacy notice.

Some models are served by third-party providers, including Lyber. Their acceptable use terms apply to your requests as well: in particular, no fraud, unlawful exploitation, unauthorized access, credential theft, interference with other customers or infrastructure, or bypassing of account, payment or capacity controls. Model output is also subject to each model's licence.

We may suspend keys or accounts to contain abuse, a security incident or a legal obligation, and we will explain why where we lawfully can. Content involving minors is reported to the authorities. Report abuse to abuse@batchwell.ai.