How NASA Is Opening Its Moon and Telescope Data — and How You Can Join In

Why this matters: NASA’s recent push toward open science

In September 2026 NASA described a coordinated effort to make Artemis program science and other datasets more openly available, while also sharing practical tools and methods for data stewardship. The agency highlighted two related actions: workshops with Artemis Accords signatories about timely scientific data release and examples of NASA’s existing public archives and tools (including the Planetary Data System); and the launch of a citizen‑science project called Artifact InSPECtor that invites volunteers to help clean telescope data. Those items are part of a broader, explicit policy stance: NASA wants datasets, analysis models and tools to be more reusable and interoperable so international partners and the public can contribute to and verify results.

What NASA actually announced (the facts)

NASA hosted two virtual workshops that focused on open science principles and tools for implementing them, aimed at technical experts across the 71 countries that have signed the Artemis Accords. The agency presented existing systems — notably the Planetary Data System — and shared practices for curating and distributing planetary and lunar data. Separately, NASA Science launched Artifact InSPECtor, a citizen‑science project that asks volunteers to identify and label imaging and spectral artifacts in data from the ESA Euclid mission and, starting in early 2027, the Nancy Grace Roman Space Telescope. The announced goals are transparency, reproducibility and practical interoperability among international partners.

Artemis Accords: a short explainer and its limits

The Artemis Accords are voluntary principles signed by governments that commit to transparent, peaceful space exploration and to enabling access to scientific data. NASA’s recent sessions focused on how signatories can adopt consistent open‑data practices. Important limits: the Accords establish principles but do not create a single, binding technical standard. Real implementation depends on each signatory building or adapting its own data‑sharing framework, and the workshops were framed as practical peer exchanges rather than a legal mandate. In short: the Accords create common ground, not an automatic, one‑size‑fits‑all data portal.

Planetary Data System and the model NASA shared

NASA used the Planetary Data System (PDS) as a working example in the workshops. PDS is a curated archive for planetary science; it defines a data information model and long‑term curation processes that make datasets discoverable and reusable. For groups aiming to share planetary or lunar data, PDS provides a mature model for metadata standards, file packaging and versioning. The practical message in the workshops was not that every participant must adopt PDS wholesale, but that interoperability and clear metadata practices are essential for third parties — researchers, educators and software developers — to reuse and reproduce results.

Artifact InSPECtor: what volunteers will do and why it helps

Artifact InSPECtor invites the public to examine real telescope spectra and images and to mark features that are likely non‑astronomical — things like cosmic‑ray hits, electronic glitches, or stray light. Volunteers learn by example and use a phone, tablet or computer to participate. Those human labels will be used to improve machine learning models that automatically flag or remove artifacts. The project initially uses Euclid data and will add Roman Telescope data when available. The value proposition is concrete: human verification can reduce false positives and teach models to generalize to new instruments, speeding up science that depends on clean spectral and imaging data.

Worked example: how a volunteer session improves a dataset

Imagine you open Artifact InSPECtor and are shown a spectrograph image from Euclid with a bright streak across several pixels. The interface teaches you what streaks from cosmic rays typically look like, then asks you to tag the region as an artifact and select a category (cosmic ray, stray light, detector defect). Your labeled example is combined with thousands of others to retrain the artifact‑detection model. After retraining, the model is less likely to mistake real emission lines for defects. This is the exact pipeline NASA describes: human labels -> improved AI filters -> cleaner inputs for scientific analysis. The approach is iterative and relies on broad participation to capture varied artifact types.

Decision checklist: should you take part?

Use this quick checklist before volunteering:

• Interest and time: Do you want to spend short sessions (minutes) learning and labeling images?
• Expectations: Are you contributing to data curation, not getting credit as an author on NASA papers?
• Privacy and accounts: Will you use a NASA or third‑party account as required, and are you comfortable with the project’s data‑use terms?
• Skill level: Artifact InSPECtor explicitly supports beginners and children, so no specialist background is required.
• Usefulness: Do you want to help machine learning models improve so they benefit later scientific analysis?

If you can answer yes to most items, this project is a fit; otherwise, consider passive ways to follow the output (public images, papers, and PDS releases).

Practical steps to join — a short how‑to

Step 1: Visit the Artifact InSPECtor project page and read the brief tutorial. Step 2: Complete the example training tasks the site provides; these teach you artifact categories. Step 3: Start labeling real data in short sessions; you can stop and resume. Step 4: Watch project updates — improved models and cleaned datasets are typically reflected in public releases or in companion papers. Finally, when using labeled or cleaned data on your own, cite the original source as directed on the project page and check licensing. All of these steps are described or implied in NASA’s public project materials.

Practical limits and risks to keep in mind

Open science is not instant. The NASA materials emphasize that technology is necessary but not sufficient: data sharing also requires cultural and procedural changes to make science reproducible. Implementation gaps include inconsistent metadata practices across signatories, legal and export controls on some datasets, delays in standardizing formats, and the need for long‑term curation resources. For Artifact InSPECtor specifically, volunteer labels improve models but cannot replace professional calibration and instrument teams; labeled datasets are one input in a larger pipeline. Be cautious about overinterpreting preliminary public data releases and check dataset provenance before reusing it in research or public claims.

What this means for researchers, educators and the curious public

For researchers: the workshops and shared examples aim to reduce friction when combining data across missions and nations, but researchers will still need to validate instrument‑specific calibrations. For educators: Artifact InSPECtor can be a hands‑on classroom exercise that illustrates how data quality matters and how human and machine work together. For the public: these initiatives expand opportunities to contribute meaningfully to active science, from labeling artifacts to exploring public archives. If you plan to reuse NASA datasets, start with the Planetary Data System model and follow any dataset‑specific documentation and citation guidelines provided with the release.

Sources

1. “NASA Boosts Open Science, Data Sharing with Artemis Accords,” NASA, Sep 11, 2026. https://www.nasa.gov/organizations/oiir/artemis-accords/nasa-boosts-open-science-data-sharing-with-artemis-accords/
2. “Help Refine Data from Space Telescopes with Artifact InSPECtor,” NASA Science, Sep 11, 2026. https://science.nasa.gov/get-involved/citizen-science/help-refine-data-from-space-telescopes-with-artifact-inspector/
3. “NASA’s Chandra Spots Galactic Gem,” NASA, Sep 11, 2026. https://www.nasa.gov/image-article/nasas-chandra-spots-galactic-gem/