Overview
On 13 February 2026, the All-domain Anomaly Resolution Office added a new information paper to its public records site: 2025 UAP Workshop: Narrative Data, Infrastructures, and Analysis — Workshop Synthesis and Recommendations. The paper arose from a two-day workshop held on 5-6 August 2025 at Associated Universities, Inc. in the Washington, D.C. area and brought together 40 participants from government, academia and independent research organisations. The workshop was sponsored by AARO, organised through a collaboration involving Associated Universities, Inc. (AUI) and Florida State University (FSU), and focused specifically on the problem of working with UAP narrative data. AARO — 2025 UAP Workshop paper AARO — UAP Records/Information Papers
The release date requires correction from some later timeline summaries. The document was cleared for open publication by the Department of Defense Office of Prepublication and Security Review on 11 February 2026; AARO's own records index lists it as added on 13 February 2026; and AUI publicly discussed the release on 19 February. It was therefore not first published in July 2026. AARO — UAP Records/Information Papers AUI — AARO Releases Report on Unidentified Anomalous Phenomena
The paper is important less for any conclusion about what UAP are than for its unusually direct description of why much existing UAP information is difficult to analyse scientifically. It characterises the data landscape as fragmented, sparse and unstructured, spanning military logs, pilot accounts, archival records, civilian testimony, online reporting systems, social media and technical sensor material. These sources are often difficult to combine because they were created for different purposes, use inconsistent terminology, omit important metadata or are restricted by classification, privacy, retention and access rules.
Its recommendations are correspondingly infrastructural. The participants argued for standard metadata templates; better preservation of original records; links between narrative reports and contextual data such as weather, air traffic and astronomy; systems capable of integrating qualitative and quantitative evidence; cautious use of AI for transcription, clustering and triage; and continued human oversight.
These are substantial methodological proposals.
They should not be mistaken for an empirical finding that AARO had discovered a new class of UAP, demonstrated extraordinary technology or validated the reports discussed by workshop participants. The paper is a workshop synthesis, not a peer-reviewed case-resolution study or statistical analysis of a defined UAP population.
Its scientific significance lies in specifying what a better evidence system might look like.
Chronology
5–6 August 2025 — workshop.
AARO sponsors an expert workshop on narrative UAP data.
13 February 2026 — publication indexed.
AARO adds the workshop synthesis paper to its official records index.
Document and purpose
The distinction between the workshop paper and an AARO case report is important.
The 17-page document synthesises discussions held during a workshop. It contains an executive summary, descriptions of three breakout sessions, a synthesis of findings, recommended next steps and appendices reproducing the invitation, participant conduct guidelines, agenda and discussion prompts. AARO — workshop paper
It does not present a controlled experiment. It does not disclose a new catalogue of UAP cases. It does not quantify the proportion of reports attributable to balloons, aircraft, satellites, sensor artefacts or unresolved phenomena. It does not provide an evidential basis for estimating how many UAP reports represent one underlying phenomenon. The output is therefore better classified as methodological and infrastructural guidance generated through expert discussion. That gives it real value, but a different kind of value from an annual report or case-resolution paper.
The workshop's conclusions reflect areas of agreement and concern among the participants who attended. They are not automatically equivalent to validated scientific consensus across the wider fields of astronomy, atmospheric science, aviation safety, information science or intelligence analysis.
The paper itself does not claim otherwise.
Participants and limits
The final workshop included 40 participants.
According to the report, participants were selected because they had demonstrated expertise in one or more fields including artificial intelligence and machine learning, UAP research and data, physical and natural sciences, information and data science, archives and records, analytical methods, cyberinfrastructure, computation, and human and social sciences. The invitation initially anticipated a smaller group of roughly 25-30 people, but the final attendance was larger. AARO — workshop paper
The workshop intentionally protected participant privacy.
The organisers obtained Institutional Review Board approval governing data collection and security. They decided not to publicise the meeting in advance, allowed participants to omit affiliations from name badges, prohibited photography without consent and instructed attendees not to attribute statements without permission. Breakout discussions could be recorded for transcription, but identifying information was to be removed and recordings destroyed after transcription and verification.
This design has a reasonable purpose. UAP remains professionally contentious, and the organisers explicitly wanted participants holding differing views to speak without concern that individual remarks would be publicly attached to them. It also creates an evaluation limitation.
The published synthesis does not provide a complete named participant list or enough information for an outside reader to assess the balance of disciplines, institutional affiliations or prior positions represented in every discussion. We can establish that government, academic and independent-research communities were present, and some participating organisations later identified themselves publicly. The Society for UAP Studies, for example, said that it attended alongside organisations including the Scientific Coalition for UAP Studies, National UFO Reporting Center and National UFO Historical Records Center. Society for UAP Studies — workshop participation
That information helps establish breadth. It does not allow the workshop to be treated as a representative survey of the entire scientific community.
Earlier workshop
The 2025 event did not emerge in isolation.
Its invitation states that the workshop built upon a 2024 meeting titled Unidentified Anomalous Phenomena (UAP): A Dialogue on Science, Public Engagement and Communication. That earlier workshop was funded through a National Science Foundation-supported ASTRO ACCEL project and was held from 15-17 May 2024 in the Washington, D.C. region. 2024 UAP Workshop Executive Summary — Zenodo
The earlier event concentrated more heavily on science communication, public engagement and the social environment surrounding UAP research.
Its executive summary identified a need for better communication, interdisciplinary cooperation and improved mechanisms for gathering and interpreting information. AUI later described advanced tools and techniques for data collection and analysis as one of the important recommendations emerging from the 2024 work. AUI — 2026 release statement
The 2025 workshop can therefore be understood as a methodological continuation. The subject moved from how science and the public should communicate about UAP toward how the reports themselves should be structured, linked and analysed. The funding and sponsorship also changed. The 2024 workshop was supported through NSF funding. The 2025 narrative-data workshop was sponsored by AARO. That distinction should be preserved. The 2025 paper is not an NSF report, even though it grew partly from an earlier NSF-supported effort.
Narrative and measurement
One of the paper's strongest contributions is the distinction it implicitly draws between the value of a witness narrative and the physical quantities that can legitimately be derived from it. Narrative reports are often the starting point of UAP investigation.
A pilot may describe the appearance, motion, duration and apparent behaviour of an object. A military log may record where a unit was operating and when an event occurred. An oral history may identify other witnesses or systems that could have produced corroborating records.
These are valuable investigative functions. A narrative alone does not necessarily establish physical distance, size, velocity or acceleration.
The workshop's proposed reporting improvements reflect this problem directly. Participants recommended asking witnesses how they estimated distance, size or speed rather than treating such values as inherently measured quantities. They also recommended framing questions around what was perceived—angular size, shape, movement, sound or effects—rather than forcing a reporter to assert an exact physical property they may not actually know. AARO — workshop paper
This is scientifically important.
An object that appears to move rapidly across the sky may be nearby and slow, distant and fast, or affected by observer motion. Without range, line-of-sight geometry and timing, apparent angular motion cannot be translated confidently into linear speed.
The paper therefore treats narrative data primarily as a source of observational claims and investigative leads, not as a substitute for calibrated sensor measurements. That is a more rigorous position than either dismissing testimony as worthless or accepting reported kinematics at face value.
Metadata
The workshop repeatedly identifies missing metadata as a central weakness in existing UAP datasets.
At minimum, a useful report should ideally preserve time, date and location with enough precision to permit external correlation. Depending upon the evidence type, additional fields may include morphology, duration, number of objects, observer position, sensor type, frequency band, platform state, weather conditions and witness background. AARO — workshop paper
Provenance receives particular emphasis. For an image or video, investigators need to know where the file came from, what device created it, whether embedded metadata remain intact, whether the file was modified and how it moved between people or systems before analysis. This is not administrative detail.
A visual file without provenance can become detached from the very information needed to interpret it. Compression, cropping, re-encoding, slow-motion processing or enhancement can alter apparent morphology and motion. A timestamp copied manually can be wrong. A location added later may reflect memory rather than sensor metadata.
The workshop's emphasis on preserving source files and automatically ingesting photo metadata is therefore one of its most practically important recommendations.
AARO's 2026 public video releases illustrate the same issue. The office repeatedly distinguishes what can be described visually from what can be concluded analytically and notes when media were digitally altered or lack complete provenance. AARO — UAP Report Documents
The workshop paper provides the infrastructure-level explanation for why those caveats matter.
Metadata classes
A particularly useful distinction in the report is between descriptive and interpretive metadata. Descriptive metadata concerns information that can be recorded comparatively directly: time, location, apparent shape, number of objects, sensor position or frequency band. Interpretive metadata concerns meaning assigned to an experience: whether an object appeared responsive, whether a witness believed there were physical effects, or how an event was experienced. The two types should not be collapsed. A description such as “a light moved from left to right over approximately ten seconds” is different from “the object reacted intelligently to my aircraft.” The second claim may be important and worth preserving, but it contains an inference about causation or intention.
The workshop's recommendation to preserve narrative richness alongside structured data is useful precisely because it avoids forcing one type into the other. A purely quantitative database can strip away details that later prove relevant. A purely narrative archive makes comparison and statistical analysis extremely difficult. The proposed solution is to keep both.
Independent context
The paper repeatedly argues that narrative evidence becomes more useful when it can be connected to other datasets.
Suggested sources include Federal Aviation Administration and NASA Aviation Safety Reporting System material, Automatic Dependent Surveillance-Broadcast flight tracks, weather radar, astronomical databases, fireball networks, satellite imagery, seismological data, CCTV and even doorbell cameras. AARO — workshop paper
This recommendation has direct evidential consequences. Suppose a witness reports a bright object moving slowly near the horizon.
A timestamp and coordinates allow investigators to test whether a known aircraft occupied that line of sight, whether a satellite flare was visible, whether a meteor or fireball was recorded, whether weather balloons were present or whether another camera captured the same event.
Without those links, the report may remain unresolved because the context needed to test ordinary explanations disappears with time. Corroboration can also strengthen unusual cases. If multiple independent sources record the same event and their measurements are mutually consistent, explanations based upon a single witness error or one sensor artefact become less likely. The paper therefore treats corroboration rather than reputation as the more defensible route to evidential credibility. That is an important principle for UAP research.
Professional observers
The workshop describes military reports and ship logs as particularly robust sources because they may contain structured information about platforms, pilots and operations. It simultaneously warns against giving excessive automatic weight to certain professions.
Participants noted that pilots, police and military personnel may have useful observational training, but an analytical system can become biased if professional status is used as a proxy for case quality without regard to the actual data collected. AARO — workshop paper
This is a subtle but important point. An experienced pilot may be better than an untrained observer at recognising aircraft behaviour, altitude relationships or cockpit phenomena. That does not provide the pilot with exact range to an unfamiliar distant object unless range was actually measured. Expertise is relevant to witness assessment, but it does not transform an estimate into instrumentation. The workshop's proposed triage systems therefore attempt to combine witness characteristics with data richness and independent corroboration rather than relying upon occupation alone.
Historical archives
The paper recommends continued preservation and digitisation of historical UAP reports. This is significant because the workshop did not advocate abandoning old cases simply because modern data collection could be better. Historical records can reveal reporting patterns, identify witnesses, document government activity and sometimes preserve technical material that remains analysable decades later. However, the report is realistic about the problems.
Older material may be handwritten, poorly indexed, stored in incompatible formats or missing contextual information. Optical character recognition struggles with cursive writing. Crowdsourced transcription can introduce errors. Translation may distort meaning. Retention failures can remove critical records entirely.
The paper cites the well-known loss of some data associated with the 2004 Nimitz encounters as an example of why retention matters. AARO — workshop paper The recommendation is consequently dual. Prioritise better contemporary collection because investigators can still control what is recorded. Preserve older material because information that already exists may contain unique historical value. This avoids two opposite errors: expecting fragmentary historical reports to meet modern scientific standards, or discarding them because they do not.
AI limitations
Artificial intelligence receives unusually detailed treatment. The workshop identified legitimate uses for AI and machine learning: transcription, extraction of dates and locations, semantic search, clustering, pattern detection, multimodal comparison and preliminary triage of reports. These applications are attractive because UAP archives contain large amounts of unstructured text and media. The paper is equally explicit about the risks.
Large language models can hallucinate. Training data already contain decades of UFO mythology, speculation and repeated claims, which can bias model outputs. Small and unrepresentative datasets can make specialised training unreliable. Generative AI can also create fake images, videos and textual reports that contaminate future datasets. AARO — workshop paper
The resulting recommendation is a human-in-the-loop model. AI can propose structured fields from free text, filter large report sets or identify candidate clusters. Humans should verify the extraction, inspect the source material and retain responsibility for contextual interpretation. This is methodologically stronger than treating AI as an automated UAP detector. A language model can identify that a narrative contains the phrase “rapid acceleration.” It cannot determine that physical acceleration occurred unless the underlying evidence supports the claim.
Data quality
The workshop summarises one limitation bluntly: AI analysis remains dependent upon input quality. This is especially relevant to historical UAP databases. If multiple websites reproduce the same original anecdote, an algorithm may mistake repetition for independent corroboration unless source dependency is tracked. If a report's date is wrong, cross-correlation with flight or astronomical data may produce a false negative. If a low-resolution image has been enhanced repeatedly, a computer-vision model may classify processing artefacts as physical features. If a dataset contains mostly spectacular reports because mundane observations were not submitted, statistical patterns will inherit that selection bias. AI does not automatically correct these problems. At scale, it can amplify them.
The workshop's emphasis on provenance, common metadata schemas and source preservation is therefore a prerequisite to the AI recommendations rather than a separate topic.
Taxonomies
The workshop proposes a concise taxonomy of familiar reported shapes such as disk, sphere and triangle while retaining an open “other” field. The purpose is practical.
Structured categories allow analysts to compare reports across databases. If one archive uses “orb,” another “sphere,” and another “round light,” a data dictionary can help determine when those categories should be considered equivalent and when they should remain distinct.
There is nevertheless a methodological risk. Taxonomies can influence what witnesses report.
If a form prominently offers “triangle,” “disk” and “cigar,” reporters may select the closest familiar category rather than describe an unusual form in their own words. Once coded, that choice may later appear as evidence that a stable “triangle class” exists.
The workshop partly addresses this by recommending that free narrative be collected before structured extraction and that unusual forms remain available through free text. That sequencing is important. A taxonomy should organise observations after collection rather than manufacture them during collection.
Unknown values
Another strong recommendation concerns how forms handle missing knowledge. The workshop proposes “refuse to answer” or equivalent options rather than forcing witnesses to provide values they do not know. This is a small design feature with significant analytical consequences. A witness who cannot estimate distance may otherwise guess because the form requires a number. That fabricated precision can later enter statistical analysis as though it were an observation. The same problem applies to exact size, altitude, speed and time. Good data infrastructure therefore needs to preserve uncertainty as data. An unknown value is often more scientifically useful than a confident-looking estimate invented to satisfy a required field.
This principle aligns closely with the way AARO's current FAQ describes useful UAP information: metadata, higher-resolution imagery and measured radio-frequency information with location and timestamps are valuable specifically because they replace guesswork with recorded quantities. AARO — Frequently Asked Questions
Classification and interoperability
Classification remains one of the paper's most difficult problems.
Military UAP observations may be captured by classified sensors. The classification can attach not because the observed object is extraordinary, but because revealing the sensor's capabilities, location or mission would expose sensitive information.
This creates an analytical barrier.
Civilian researchers may possess relevant weather or astronomical data but lack access to the military observation. Government analysts may possess classified sensor records but have difficulty sharing enough detail for external replication.
The workshop proposes interoperable infrastructures that can link information while respecting classification, privacy and ethical constraints. That is easier to recommend than implement. The document does not specify a finished technical architecture capable of bridging classified and unclassified systems. It instead recommends pilot-scale integration projects, data dictionaries, modular metadata standards, interface-control documents, APIs and governance structures. These are practical starting points rather than a complete solution. The significance of the recommendation is that information architecture itself is treated as part of the scientific problem.
AARO casework
The workshop paper becomes more useful when compared with AARO's annual reporting.
AARO's FY2025 report states that a lack of timely and actionable sensor data continues to constrain case resolution. It says the office has been working with military and technical partners to define optimal sensor requirements, information-sharing processes and the essential elements needed for high-quality UAP reporting. AARO — FY2025 Consolidated Annual Report
The same annual report provides a useful example of what narrative data can and cannot do.
AARO says that FAA-derived civilian pilot reports are less data-rich than many Department of War reports but can still contain enough narrative detail to support high-confidence assessments, particularly where descriptions are consistent with satellite flaring.
Conversely, AARO reports receiving narratives concerning phenomena near sensitive infrastructure whose alleged characteristics, if validated, might exceed known performance. Those reports lacked accompanying technical data, preventing the alleged performance from being established. AARO — FY2025 Annual Report
This closely matches the workshop's methodological position. Narrative evidence can sometimes be sufficient to identify a familiar phenomenon. Narrative evidence is much less able to substantiate extraordinary physical performance without technical corroboration.
Existing standards
The paper should not be interpreted as though AARO had no reporting structure before 2025.
The FY2023 annual report states that AARO, working with the Joint Staff and Department of Defense components, had already led development of UAP reporting standards and was working to improve data quality from military sources. AARO — FY2023 Consolidated Annual Report
AARO also published a detailed user guide in 2023 for its authorised historical programme-reporting mechanism, while operational military reporting proceeded through separate classified instructions. AARO — UAP Program Report User Guide
The 2025 workshop therefore represents an expansion of the data problem rather than its discovery. Military operational reporting can be standardised internally. The harder problem is linking military reports to historical archives, civilian narratives, scientific sensor data and independent research databases that were created under different schemas. The workshop's focus was this larger ecosystem.
Institutional position
The workshop's sponsorship should be acknowledged in assessing the paper. AARO funded the meeting, and the final paper is hosted as an AARO information paper. AUI and FSU organised the event and produced the synthesis. This makes the document highly relevant to AARO's developing research approach. It also means the exercise should not be presented as an independent external audit of AARO's existing data practices. Participants may have criticised barriers and shortcomings, and the paper plainly acknowledges many. But the workshop was designed in collaboration with the organisation whose mission it was intended to support. The distinction is similar to other government-sponsored technical workshops.
The output can contain serious scientific analysis without possessing the same independence as a study commissioned specifically to audit programme performance. AUI's own release describes the report as a collaboration between AARO, AUI and FSU rather than an independent verdict on AARO. AUI — report release
Evidential limits
The document is unusually easy to overinterpret because it uses language associated with scientific research, AI and large-scale data analysis. Its recommendations do not demonstrate that statistical patterns have already been found in UAP narratives. They do not establish that historical morphology categories correspond to distinct physical craft. They do not establish that AI can reliably distinguish anomalous from conventional reports. They do not validate the authenticity of databases maintained by independent UAP organisations merely because those sources were discussed. They do not show that narrative reports can establish extraordinary kinematics without sensor data. The paper is a roadmap for making those questions more testable. That is a meaningful contribution without being a discovery.
Implementation status
The most important question after publication is whether the recommendations have been implemented. The public record shows partial alignment rather than a clearly identifiable complete implementation programme.
AARO's FY2025 annual report, released in July 2026, says the office continues to refine processes and adopt new analytical tools. It describes three-dimensional modelling used to resolve large numbers of satellite-flare reports and continued work to improve sensor requirements and information sharing. AARO — FY2025 Annual Report
AARO's public site also now provides significantly more mission reports, imagery and video than was available when the workshop occurred. The office frequently publishes contextual narrative descriptions alongside media and carefully distinguishes description from investigative conclusion. AARO — UAP Report Documents
Those developments are consistent with parts of the workshop's emphasis on preservation and contextualisation. They do not establish that a common cross-organisational metadata schema, integrated civilian-military data infrastructure or public narrative-data API has been implemented.
As of 26 August 2026, AARO's public information-paper index does not identify a follow-up implementation report specifically documenting completion of the workshop's recommended metadata templates, interoperability pilots or AI-governance framework. AARO — UAP Records/Information Papers
The paper should therefore still be treated principally as a recommendation document.
Historical significance
The paper marks an important shift in the institutional UAP discussion. Earlier government programmes often focused principally upon resolving individual reports. The 2025 workshop asks a more fundamental question: what must the data system look like before large-scale UAP analysis can be trusted? That shift matters.
A case-resolution programme can repeatedly encounter the same weaknesses without fixing their source. If observers are not asked the right questions, if sensor files lose provenance, if archives cannot communicate, if terminology is inconsistent and if AI systems are trained on duplicated or culturally biased material, then more analysis may simply reproduce the same uncertainty more efficiently.
The workshop treats those weaknesses as research problems in their own right. It also brings information science, archival studies and social-science methods into a field usually dominated by aviation, physics and intelligence. That interdisciplinary expansion is appropriate because much of the UAP problem is not initially a physics problem. Before a physicist can calculate acceleration, someone must preserve the timestamp, range and sensor geometry. Before a statistician can identify a cluster, someone must determine whether repeated reports are independent. Before an AI system can classify a narrative, someone must define the taxonomy and preserve the original text. The 2025 workshop makes those dependencies explicit. Historically, that may prove more important than any single proposed AI tool or reporting form.
It reframes UAP research as an evidence-infrastructure problem before it becomes an explanation problem.
Evidence assessment
The 2025 narrative-data workshop paper is one of AARO's more methodologically useful public documents because it addresses weaknesses that can distort UAP research regardless of what the underlying phenomena ultimately prove to be.
Poor timestamps are a problem whether an object is a satellite or genuinely anomalous. Missing provenance is a problem whether an image depicts a balloon or an unknown aircraft. Selection bias is a problem whether a dataset ultimately contains zero extraordinary cases or several. AI hallucination and duplicated sourcing can generate false patterns under any hypothesis. The paper therefore contributes something more durable than a claim about one case. Its strongest recommendations concern how evidence should be preserved before interpretation begins. Free narrative should be retained because structured categories can erase unusual details. Structured metadata should be added because narrative alone is difficult to compare. Uncertainty should be recorded rather than replaced by guessed precision.
Independent contextual data should be linked wherever possible. Original media and provenance should be preserved. Automated methods should assist human analysts rather than replace evidential judgement. These principles are consistent with mainstream scientific and information-management practice. The paper's limitations are equally important.
It is a synthesis of a deliberately small, selective workshop rather than a peer-reviewed empirical study. The complete participant composition and underlying anonymised deliberations are not available for independent analysis. The report does not test whether its proposed infrastructure actually improves UAP classification accuracy. Nor does it specify quantitative benchmarks by which future implementation should be judged.
The practical importance of the paper therefore depends upon what happens next. A common schema is valuable only if reporting systems adopt it. Provenance standards matter only if original files are actually retained. AI safeguards matter only if automated analysis is evaluated against known cases and error rates are measured. Interoperability matters only if legal, security and technical barriers can be overcome sufficiently for datasets to be linked. As of August 2026, the public record shows AARO continuing to improve reporting and analytical tools, but it does not yet demonstrate a complete implementation of the workshop's proposed data ecosystem. The appropriate assessment is therefore positive but bounded. The workshop paper does not solve UAP.
It provides a comparatively credible description of why the available evidence so often fails to support confident conclusions—and what would need to change for future research to become more scientifically informative.
null
AARO — 2025 UAP Workshop: Narrative Data, Infrastructures, and Analysis
The principal source for this deep dive. The 17-page paper contains the workshop methodology, discussion summaries, synthesis, recommendations, invitation, conduct rules, agenda and breakout prompts. It establishes the 40-participant attendance, the focus on narrative data, the metadata and provenance recommendations, the proposed human-AI workflow and the limitations identified across current UAP data systems.
Read the official AARO workshop paper
AARO — UAP Records / Information Papers
The controlling AARO publication index for the document. It lists the workshop paper as new content added on 13 February 2026, correcting later timeline summaries that place its initial release in July.
View AARO's UAP Records index
Associated Universities, Inc. — “AARO Releases Report on Unidentified Anomalous Phenomena (UAP),” 19 February 2026
Primary organiser statement describing the collaboration between AARO, AUI and Florida State University and connecting the 2025 workshop to the earlier NSF-funded 2024 effort. Useful for establishing the organisational relationship and the intended emphasis on rigorous data collection, standardisation and analysis.
Read the AUI release
2024 UAP Workshop Executive Summary — Unidentified Anomalous Phenomena: A Dialogue on Science, Public Engagement and Communication
Primary workshop summary deposited on Zenodo by the organisers. Documents the May 2024 NSF-supported predecessor meeting and its focus on science communication, public engagement and interdisciplinary UAP research.
Read the 2024 workshop executive summary
Florida State University — June 2024 faculty and staff record
Contemporary university record confirming that FSU's Gretchen Stahlman co-led the 2024 UAP workshop in the Washington, D.C. area with National Science Foundation support.
Read the FSU record
Society for UAP Studies — participation in the August 2025 workshop
Participant-organisation account identifying its involvement and several other independent research organisations represented at the workshop. Useful for adding limited visibility to a participant list that the official paper intentionally anonymises. It is a participant position source rather than an independent evaluation of the workshop's findings.
Read the SUAPS workshop account
AARO — Fiscal Year 2025 Consolidated Annual Report on UAP
Important follow-up evidence for evaluating the workshop against AARO's operational casework. The report states that insufficient timely and actionable sensor data continue to constrain resolution, while describing improvements in sensor requirements, information sharing and analytical modelling. It also demonstrates circumstances in which narrative reports can support conventional attribution and circumstances in which extraordinary-sounding narrative claims remain unvalidated without technical data.
Read AARO's FY2025 Annual Report
AARO — Fiscal Year 2023 Consolidated Annual Report on UAP
Earlier official evidence that AARO and the Joint Staff had already begun developing operational UAP reporting standards before the 2025 workshop. Relevant for distinguishing the workshop's broader cross-dataset infrastructure recommendations from the existence of pre-existing military reporting procedures.
Read the FY2023 Annual Report
AARO — U.S. Government UAP-Related Program/Activity Report User Guide
AARO's 2023 guidance for its authorised historical programme-reporting mechanism. Demonstrates that the office already used structured forms and eligibility rules for one class of narrative evidence while maintaining separate procedures for current operational UAP reporting.
Read the AARO reporting user guide
AARO — Frequently Asked Questions / scientifically useful UAP reporting
Current AARO guidance describing the value of metadata, higher-resolution imagery, timestamps, location and measured radio-frequency information. Useful for comparing the workshop's recommendations with the office's present public guidance on evidentially useful reports.
Read AARO's FAQ
AARO — UAP Report Documents
Current public repository of AARO-released mission reports, imagery and video. Particularly relevant to the workshop's concerns about provenance, contextual descriptions and the distinction between observable image content and analytical conclusion.
Browse AARO's UAP Report Documents
National Archives — UAP/UFO bulk downloads
The current federal archival infrastructure for large-scale access to digitised historical UAP/UFO government records. Relevant to the workshop's recommendation that historical material be preserved and digitised while newer reporting systems prioritise better structured data.
Browse NARA UAP/UFO bulk downloads