Got questions? We've got answers.
Here's everything you need to know before getting started.
General Questions
Here's everything you need to know before getting started.
Pretense sits between your developers and the AI tools they use, like a proxy on the machine. Before any code or prompt reaches an LLM, it finds the sensitive parts and swaps them for realistic stand-ins. The AI works on the safe version, and when the response comes back, Pretense restores the original values. Your real data never leaves the machine.
The sensitive parts don't. Credentials, customer data, patient records and proprietary identifiers are swapped for stand-ins before anything is sent, so the model only ever sees the masked version. The real values are restored locally when the response returns.
Traditional DLP was built to watch email and file uploads. It was never built for a developer pasting code into ChatGPT or Claude at 2am, so that traffic passes straight through it. And when DLP does catch something, it blocks or redacts, which breaks the AI's answer. Pretense is built for the AI workflow specifically, and it protects the data without degrading the output.
No, and that's the whole point. Other tools redact or delete the sensitive parts, which strips the context the model needs and makes the answers worse over a session. Pretense swaps values for stand-ins that keep the same structure, so the AI never knows anything is missing and the answer is as good as if you'd sent the real thing.
Pretense detects the sensitive parts before they reach the AI: credentials, API keys, PII, patient records, SSNs, and proprietary source. It swaps each one for a realistic stand-in that keeps the same structure, so the code still works and the AI can still reason about it. Once the response comes back, the real values are restored on your machine.
On your machine. Pretense runs locally as a proxy, so the sensitive values are swapped before anything is sent. Your real data is never processed on our servers.
No. Developers write code and use their AI tools exactly as they do now. Pretense runs in the background as a proxy, with no change to the editor, the model, or how anyone works. Setup is a one-time install.
About 30 seconds. You install the CLI with one command, create your organization, and generate an API key. That's it, you're protected. No procurement process and no rebuilding your stack.
Yes. You create an organization, and activity, access and usage are tracked across everyone on the team. Paid plans add more seats, longer log retention and admin controls. Free supports up to 3 seats; Pro supports 25+.
Audit trails are available today. Formal SOC 2 and HIPAA certification are in progress and listed as coming soon on our Enterprise plan.
No. The Free plan lets you start protecting code without a card. You only add billing if you upgrade to Pro or Enterprise.
You can change plans at any time. Upgrading unlocks more seats, longer log retention and priority support. If you cancel, you keep access through the end of your billing period and then move back to the Free plan. Your setup stays in place.