Are AI model outputs considered ITAR or EAR controlled technical data?

It depends on the content and the model's purpose: outputs directly tied to USML defense articles can be ITAR technical data, while dual-use outputs may fall under the EAR (EAR99 or a CCL classification), and much general output is uncontrolled —…

register 09 · Compliance pins· ITAR
01 ·

Answer.

Export control · ITAR.

It depends on the content and the model's purpose: outputs directly tied to USML defense articles can be ITAR technical data, while dual-use outputs may fall under the EAR (EAR99 or a CCL classification), and much general output is uncontrolled — classification is fact-specific. Recent AI-diffusion rules also reach certain model weights and compute. Planisphere does not classify your data under ITAR/EAR; it keeps the audit in-boundary so you can run it without an unintended export, and the legal classification stays with your export-control counsel.

02 ·

The mark behind the answer.

ITAR draws a boundary around technical data and who may receive it. Planisphere's posture is the answer: model state nev…

USML technical-data control · deemed-export boundary for model artifacts.

→ Full reference for ITAR

03 ·

More on ITAR.

Other questions this mark answers.

Try ITAR with a free test key.

Start with the free sandbox. Planisphere measures model behaviour and emits a reproducible, sha-pinned record — it does not certify, file, or give legal advice.

API ·

When ITAR asks for proof, hand over records — not assurances.

record the duty · seal the receipt · verify offline

Planisphere records each duty event — an output marked, a disclosure shown, a review made — and seals it into a receipt that verifies offline against our published keys. You send hashes, never content. A record is evidence a third party can check; it is not a certification and not a legal determination.

See how a record is checked