Honest answers, stated plainly.

No. Vaelon is at an early technology readiness stage: ground-verified only, no flight heritage, pre-revenue. We say this plainly because it's true, not as a formality.

Most software is tested: you run it on a set of examples and check the results. Formal verification is different. It's a mathematical proof that a property holds for every possible input, not just the ones you thought to try. That's the bar we hold the core safety checks to.

It sits between an autonomous system's decision-maker and the hardware it controls. For every proposed action, it checks a fixed set of safety rules and returns allow, block, or escalate. If it blocks something, it substitutes a safer action instead, and writes the decision into a permanent record.

It isn't an autopilot. It doesn't decide what to do, only whether a proposed action is safe. It doesn't authenticate who sent a command, verify that a sensor is telling the truth (only that its reading is physically plausible), or replace ground mission planning. It assumes it's wired directly into the command path, so a refusal only works if nothing downstream can ignore it.

No. This is not flight certification, a safety approval, or a certification of any kind. No regulator, certification body, or space agency has reviewed or approved it.

Yes. We can share a self-contained evaluation kit for your own data, see Resources, and we're happy to walk through the product directly.

We're running early evaluation pilots with outside teams on a bounded basis. Results are confidential until both sides agree in writing to make anything public, so we won't name a pilot before that happens.