For much of the twentieth century, strategic information superiority depended heavily on access. Reconnaissance satellites, interception systems, diplomatic reporting networks and specialised collection platforms allowed states to observe activity that remained invisible to almost everyone else. That hierarchy has not disappeared, but it is not the only hierarchy that matters.
Commercial satellite imagery can reveal deployments and infrastructure changes. AIS and ADS-B data can help reconstruct maritime and aviation activity. Social platforms provide eyewitness footage, while public registries, digital archives and geospatial tools expose relationships that once required specialised access to identify. Artificial intelligence can now search, translate and classify this material at a scale beyond unaided human analysis.
Observability itself has been partially redistributed. Strategic information now circulates through a mixed environment of public, commercial and classified sources. Advantage, therefore, depends not only on possessing information others cannot access, but also on recognising what matters within information available to many.
This offers a useful framework for understanding contemporary OSINT. Open-source intelligence is an assessment produced in response to an intelligence requirement. Collection is only the beginning. Verification, corroboration, contextualisation and analysis turn an observation into something that can support a judgment with an appropriate confidence level. A Telegram video, for example, acquires analytical value only after questions of geolocation, timing, provenance, recycling and independent corroboration have been addressed.
Intelligence under Conditions of Saturation
Many intelligence systems were designed around scarcity. Important information was difficult to obtain, so collection capability generated advantage. Contemporary crises often present the inverse problem. Thousands of videos, satellite images, tracking records, and competing claims can appear within hours. The constraint is increasingly the ability to process information without losing provenance, context, or analytical discipline.
Information abundance has not erased asymmetry. It has shifted where that asymmetry now sits. One institution may possess a large and sophisticated data repository but weak entity resolution, the ability to determine when different names, accounts, or identifiers refer to the same actor. Another may work with fewer sources yet combine regional expertise, source evaluation, and network analysis more effectively. The difference here is sense-making capacity.
This environment can itself be manipulated. Recycled imagery, coordinated accounts, false geolocation and selectively edited footage can saturate workflows, while algorithmic amplification rewards virality rather than relevance. Rigorous OSINT therefore requires alternative hypotheses, separate judgments about source reliability and information credibility, and a willingness to ask what evidence would disprove the preferred assessment. The observable layer is not the same as reality in full. Nor should the absence of open-source evidence be treated as evidence of absence.
Ukraine and the Narrowing of Opacity
The Russia-Ukraine war made this transformation unusually visible. Before the February 2022 invasion, commercial imagery, technology data and social-media analysis helped reveal Russian force concentrations and evolving posture outside classified channels. UK Strategic Command later described the fusion of these sources as a contribution to anticipatory intelligence, while stressing that open and secret sources were most valuable when used together.
After the invasion, satellite imagery converged with geolocated footage, Telegram posts, aircraft and vessel tracking, digital archiving and crowdsourced verification. The Centre for Information Resilience’s Eyes on Russia project, supported by a wider community including Bellingcat and GeoConfirmed, had mapped more than 11,600 verified videos and images by July 2023. Its importance was methodological as much as quantitative. Material was geolocated or chronolocated, archived, reviewed and converted from a social-media stream into a structured evidence base.
Yet independent investigators did not obtain the complete intelligence picture available to governments. States could still fuse open material with HUMINT, SIGINT, classified GEOINT, diplomatic and operational reporting, and protected databases. Public observation might establish that a unit moved without revealing its orders, readiness, or political intent. OSINT has not made the battlefield transparent. It has made opacity more difficult to maintain.
Open and Classified Intelligence as One Architecture
Treating OSINT and secret intelligence as competitors is the wrong starting point. The stronger institutional model is integration. Open information can confirm activity reported by another source, challenge an assumption, reveal anomalies or generate new collection requirements. This is the logic of collection cueing. If commercial imagery, transport data or social reporting indicates unusual activity, scarce classified assets can be directed towards a narrower intelligence question.
Institutional developments reflect this logic. The U.S. Intelligence Community’s OSINT Strategy for 2024–2026 prioritises coordinated data acquisition, integrated open-source collection management, innovation and a specialised workforce. In the United Kingdom, the National Centre for Geospatial Intelligence is the defence lead for OSINT. NATO’s Alliance Persistent Surveillance from Space initiative is building Aquila, a virtual constellation drawing on national and commercial surveillance satellites. These developments do not suggest proprietary collection is obsolete. They show external information being absorbed into formal intelligence architectures.
The New Intelligence Divide
This produces a second intelligence divide. The older divide separated actors with sophisticated collection systems from those unable to observe inaccessible targets. That divide remains. But another now separates institutions capable of converting information saturation into reliable judgment from those that merely accumulate data.
The new divide depends on technical infrastructure, commercial-data access, computational capacity and verification workflows. It also depends on assets that are harder to scale. These assets are language ability, regional knowledge, institutional memory, and analytical culture. Speed is valuable only when sustained across the entire intelligence cycle. Detecting an indicator quickly is of limited value if analytical latency, the delay between collection, validation, assessment and delivery, consumes the advantage.
It also changes the composition of intelligence work. Contemporary OSINT increasingly requires political analysts, software engineers, data engineers, geospatial specialists, linguists and data scientists inside the same production system. Technical specialists need to understand collection requirements, while analysts need enough technical literacy to recognise model limitations, data-quality problems and hidden assumptions.
AI, Verification and Political Power
Artificial intelligence accelerates this convergence. It can classify imagery, transcribe recordings, translate multilingual sources, extract entities, resolve identities and triage large document collections. Properly used, it reduces the time spent locating potentially relevant material and allows analysts to concentrate on interpretation.
But AI lowers both the cost of finding patterns and the cost of manufacturing false ones. Synthetic imagery, cloned voices, automated influence operations, and artificial personas can contaminate collection streams, while generative systems can produce fluent but hallucinated analytical outputs. As discovery becomes more automated, provenance, corroboration, and explicit confidence judgments become more, not less, important. What works in practice is an analyst supported by AI, not one replaced by it.
These changes matter politically because epistemic advantage and narrative advantage increasingly interact. An institution may possess superior classified information yet be unable to disclose the sources behind its assessment, while an external actor can publish a rapid, visually persuasive interpretation before formal analysis is complete. Traditional communication cycles now operate inside an information environment whose tempo is partly set by commercial providers, journalists, researchers, online communities and automated systems.
Technology has not removed information advantage from politics. It has changed the architecture through which that advantage is produced. Exclusive access remains powerful, especially where intentions, covert relationships and protected capabilities are concerned. But strategic superiority increasingly rests on something more organisationally demanding. It is an ability to combine fragmented sources, verify them under pressure, distinguish visibility from completeness, and deliver qualified judgment while it can still influence a decision.
What matters is the capacity to turn information into intelligence, and intelligence into institutional understanding that can still shape decisions in time.
Sources and Further Reading
U.S. Intelligence Community Open Source Intelligence Strategy 2024–2026 — CIA / ODNI: https://www.cia.gov/stories/story/ic-osint-strategy-rollout/
How open-source intelligence has shaped the Russia-Ukraine war — UK Strategic Command: https://www.gov.uk/government/speeches/how-open-source-intelligence-has-shaped-the-russia-ukraine-war
National Centre for Geospatial Intelligence - UK Government: https://www.gov.uk/government/groups/national-centre-for-geospatial-intelligence-ncgi
NATO’s approach to space / APSS-Aquila - NATO: https://www.nato.int/en/what-we-do/deterrence-and-defence/natos-approach-to-space
Over 500 Days of the Russia-Ukraine Monitor Map - Bellingcat / CIR: https://www.bellingcat.com/news/2023/07/24/over-500-days-of-the-russia-ukraine-monitor-map/
