BRIGHT EVIDENCE PACK / Experimental
A twelve-hour earlier clue to where the Sun may change.
NASA reports that a COFFIES machine-learning model can predict the appearance of solar active regions, including far-side emergence, up to 12 hours before they appear.
Canonical Bright record · JSON evidence pack · Key-facts embed
Dates and assessment
- Source published
- 2026-08-14
- Bright published
- 2026-09-19
- Substantive update
- None recorded
- Evidence state
- Experimental
- Independent verification
- Not established by this source review
- Last source review
- 2026-09-19
The claim in context
The human problem
Space-weather forecasters need earlier evidence about regions that may later produce disruptive activity.
The prior constraint
Active regions can emerge out of view or before visible surface signs become clear.
AI’s actual role
The model learns precursor patterns associated with active-region emergence.
The documented result
NASA reports up to 12 hours of lead time. The public account does not provide operational validation, false-alarm rates, or evidence of forecast-center adoption.
Why it may matter
Even a modest additional window could become another input for forecasters if performance holds under operational evaluation.
Limitations
- The model predicts active-region emergence, not a damaging flare or coronal mass ejection.
- No public false-alarm or operational validation rate is provided.
- The evidence is an agency report pending underlying paper-level performance review.
Original evidence
- NASA’s COFFIES Uses AI to Predict Storm-Causing Active Regions · government
- COFFIES Drive Science Center · government
Attribution
Credit Bright AI Future and link the canonical Bright record.
- Link to the canonical Bright record.
- Keep material limitations with the claim they qualify.
- Link to the original evidence when repeating a substantive claim.
- Do not describe a source check or organization-reported result as independent verification.
Linked source material, quotations, trademarks and media remain subject to their owners’ terms. No reuse right is granted for third-party media.