A
Layer A — Detection
Who is retelling
DISARM-aligned narrative monitoring across 7 platforms. Coordination signals: synchronized posting, content cloning, URL co-amplification, bot velocity. Every retelling tagged against DISARM Red Framework v2.
Built in Ukraine · Operational since 2024
We watch the retellers.
We measure their narratives.
retelnist · RU-UA-COG · live
IO Pressure
0.88
peak V-score
Shift Events
14
last 40 days
High Severity
6
detected
IO Propagation Pressure · 40 days▲ high alert
day 1day 20day 40
Narrative
"Kyiv forces are deliberately targeting civilian infrastructure to discredit Russia"
Recent shift events
Jun 09Coordinated amplification spikehigh
Jun 05Cross-platform polarity fliphigh
Jun 01Organic engagement uptickmedium
retelnist · UA-POP-BELIEF · weekly
Belief Δ
+9pp
vs prev month
Survey pts
7
KIIS cross-ref
Trend
↑ rising
acceleration
Belief index · rolling 40 days↑ rising trend
day 1day 20day 40
Narrative
"Western support for Ukraine is waning and will collapse by end of year"
Recent shift events
Jun 08Survey realignment detectedhigh
Jun 03Belief plateau broken upwardmedium
May 28Cross-source corroborationmedium
retelnist · ACTOR-NETWORK · snapshot
Actors tracked
38
active nodes
Clusters
5
distinct networks
New actors
3
last 7 days
Actor activity · 40 days● 3 new nodes
day 1day 20day 40
Narrative
"NATO expansion caused the conflict — Russia had no choice but to respond"
Recent shift events
Jun 10New amplifier cluster emergedhigh
Jun 06Cross-network coordination linkhigh
Jun 02Dormant actor reactivatedmedium
Retelnist is the cognitive intelligence platform that pairs DISARM-aligned detection of coordinated information operations with continuous measurement of actual effect on target population to distinguish noise from a national security event.
Methodology aligned with the DISARM Framework, EEAS FIMI taxonomy, and NATO ACT cognitive superiority doctrine.
Discipline
Boutique threat intelligence
Layer A
DISARM-aligned detection
Layer B
Belief-shift measurement
Infrastructure
EU-sovereign by default
retelnist · RU-UA-COG · live
IO Pressure
0.88
peak V-score
Shift Events
14
last 40 days
High Severity
6
detected
IO Propagation Pressure · 40 days▲ high alert
day 1day 20day 40
Narrative
"Kyiv forces are deliberately targeting civilian infrastructure to discredit Russia"
Recent shift events
Jun 09Coordinated amplification spikehigh
Jun 05Cross-platform polarity fliphigh
Jun 01Organic engagement uptickmedium
retelnist · UA-POP-BELIEF · weekly
Belief Δ
+9pp
vs prev month
Survey pts
7
KIIS cross-ref
Trend
↑ rising
acceleration
Belief index · rolling 40 days↑ rising trend
day 1day 20day 40
Narrative
"Western support for Ukraine is waning and will collapse by end of year"
Recent shift events
Jun 08Survey realignment detectedhigh
Jun 03Belief plateau broken upwardmedium
May 28Cross-source corroborationmedium
retelnist · ACTOR-NETWORK · snapshot
Actors tracked
38
active nodes
Clusters
5
distinct networks
New actors
3
last 7 days
Actor activity · 40 days● 3 new nodes
day 1day 20day 40
Narrative
"NATO expansion caused the conflict — Russia had no choice but to respond"
Recent shift events
Jun 10New amplifier cluster emergedhigh
Jun 06Cross-network coordination linkhigh
Jun 02Dormant actor reactivatedmedium
LIVE SIGNAL
Measuring belief-shift against a counterfactual
Expressed stance is what real readers voice about a tracked proposition — extracted from their reactions, comments, and posts. The counterfactual is the same audience's projected stance had the information operation never happened, extrapolated from the pre-IO trend. The gap between the two lines is the measured shift — V(x,t) — in stance units.
Narrative
"Western biotech and government laboratories are secretly engineering pathogens and bioweapons"
Expressed stance (with IO)Counterfactual (no IO)Measured shiftHigh shift eventMedium shift
Tracking window
60 days
Peak V-shift
0.34
max |stance − counterfactual|
Significant shift days
25
CI excludes zero
Counterfactual coverage
28
days with baseline
How to read this
IO is lifting the audience toward the narrative — stance sits above where the counterfactual predicts it would be without pressure.
Mean shift 0.10 across 28 counterfactual days. Confidence interval excludes zero — the effect is statistically significant.
The gap
Every other detection platform leaves this unanswered
What is the impact of the operation?
Detection tools tell you a coordinated operation is happening. That's necessary — and it's not enough.
The operations shift what populations believe, doubt, and decide.
Retelnist reveals and measures the operations.
B
Layer B — Effect
Whether the retelling is working
V(x,t) score — a continuous measurement of how belief is moving against organic baseline: frequency, sentiment polarity, identity coupling, amplification asymmetry. When the retellers push, you see the needle move.
∑
CWPI — The verdict
One number for decision-makers
Cognitive Warfare Presence Index combines both layers with statistical lag analysis. Reported with 95% confidence intervals. Bootstrap resampling. Every number carries its uncertainty.
Who uses Retelnist
Three buyer profiles
National StratCom Units
Know whether the retelling is working this week
You already know the adversary is pushing narratives. You need to know whether the retelling is working this week, in which regions, against which narratives, when the shift began. Weekly briefs in EEAS-FIMI format, ready for your director's desk.
- Weekly EEAS-FIMI briefs
- Regional belief-shift tracking
- CWPI per narrative
EU / NATO Procurement
Report in your language, with auditable methodology
You operate inside doctrinal frameworks. You need tools that report in your language with auditable methodology and reproducible results. Retelnist exports DISARM-tagged threat briefs, STIX 2.1 bundles for FIMI ISAC sharing, and statistical reports your analysts can defend.
- STIX 2.1 export
- DISARM-tagged briefs
- Permutation-tested results
Fact-checking Networks
Working tool that respects your funders' standards
You see dozens of suspicious retellings a week. You need a tool that respects your funders' standards, DISARM-compliant, EU data sovereignty, transparent methodology, exportable. Analysts can manually correct AI suggestions.
- EU data sovereignty
- Analyst corrections
- EDMO-compatible exports
One retelling, traced end to end
Ukraine deployment · anonymized scenario
Illustrative scenario from active monitoring. Every number Retelnist reports carries its uncertainty, and the methodology is documented to the level your procurement team can defend.
- Step 01
Trigger
Layer A detects 47 accounts coordinated within a 72-hour window pushing variants of "Mobilization is unjust — rich buy out, poor die."
Signal
DISARM: T0049 (Flood), T0048 (Harass), T0091 (Recruit Bad Actors)
- Step 02
Effect measurement
Layer B shows V-shift of +2.3σ on proposition P-UA-12 ("Mobilization is unjust"), with identity coupling spike on the marker "fear of ТЦК."
Signal
Belief-shift confirmed on target proposition
- Step 03
Coupling analysis
18-hour lag between Layer A peak and Layer B shift. Permutation test: p < 0.01 shows the retelling's temporal precedence is consistent with influence.
Signal
Temporal precedence established · p < 0.01
- Step 04
Verdict — CWPI
Confirmed cognitive operation with measurable impact. Brief auto-generated in EEAS-FIMI format. Counter-narrative team notified.
Signal
95% CI · bootstrap resampled
- Step 05
Closed loop
A calibrated training scenario on this exact retelling pattern is published into the inoculation module. Same propositions re-measured weeks later in trained vs untrained cohorts.
Signal
Talantir inoculation engine · McGuire 1961 methodology
The closed loop.
A platform that only detects gives you reports. A platform that only trains gives you hope. Retelnist pairs them — as a measurement system. Detect the retelling. Train the population on the technique. Re-measure the V-shift. You get statistical evidence that your program is building resistance, not just running courses.
End-to-end process
From first signal to verified report
Every detection passes through six stages — ingest, coordination detection, narrative management, belief-shift measurement, coupling significance, and reporting. Each transition is logged, outputs are auditable.
We pull raw content from 7 monitored surfaces, via content fingerprint and tier before any analysis begins. Government-relevance tags applied at this stage.
Twitter/X · Telegram · YouTubeReddit · Facebook · RSS · LinkedIn
Outputs
- Normalized content corpus · language-tagged · gov-interest flagged
- Deduplication fingerprint index for near-duplicate collapse
Closed loopThe loop closes at training.
Detection feeds measurement before the brief. The brief provides a training scenario to evaluate. V(x,t) measurement on trained vs untrained cohorts. We get statistical evidence that the program is building resistance.
V-shift re-measured post-trainingTechnique recurrence trackedResistance delta reported
Our positioning
What Retelnist is, what it is not
Retelnist is built for teams who need more than a coordination alert. Aligned with the frameworks you already operate and researching what those frameworks don't cover yet, ahead of the field. Where others stop at detection, Retelnist already pairs detection with measured effect.
They have
Detection only
- Graphika
- Logically
- Blackbird
Know that an operation is happening. Essential — not sufficient.
Retelnist adds
⊕ pairs bothDetection + Effect
- Continuous V(x,t) belief-shift measurement
- Statistical coupling — behavioral → belief
- EU-sovereign deployment by default
- Operational embedding in Ukraine
- Inoculation training with measurable outcomes
Know whether the retelling is working. The only platform that answers both questions.
Deployment & pricing
Pilot engagement
3-month structured pilot. Platform access, analyst onboarding, weekly briefs, methodology review.
Full deployment
Annual subscription per deployment + analyst seats. Volume terms for multi-deployment buyers.
Academic / NGO
Discounted pricing for verified non-commercial use. EDMO hubs and GFCN-affiliated organizations.
What you can verify
Methods documented to the level your team can defend
- DISARM Red Framework v2 — open-source, governance-shared
- DISARM tagging logic — published rules, auditable per-tag
- Bootstrap resampling N=1000, permutation tests, 95% CI on every value
- Audit trail — every tag, shift, and report logged with timestamp
- Coupling analysis methodology — published in technical brief
Not everything is open:
A cognitive intelligence platform has two failure modes. Publish all weights, and the adversary trains against your methodology. Publish nothing, and you become a black box no procurement officer can defend. Everything that can be open without arming the adversary, is open. Everything else is documented under NDA, in detail, for verified buyers.
Most cognitive security platforms are built by US/UK analytics firms watching the war from outside. Retelnist is built inside it.
Questions
The things buyers ask first
What is the V(x,t) score?
A continuous measurement for belief movement, where x is a tracked proposition and t is time. It combines frequency, sentiment polarity, identity coupling, and amplification asymmetry into a single value measured against an organic baseline and estimates belief shift. Every reported value carries a 95% confidence interval from bootstrap resampling.
What is CWPI?
The Cognitive Warfare Presence Index is a single number combining Layer A (behavioral activity) and Layer B (belief shift) with statistical lag analysis between them. It answers one question: Is this a coordinated operation that is actually working — or noise?
How does the coupling analysis work?
We model the lag between Layer A peak activity and Layer B belief shift. Permutation tests establish whether the behavioral push is causing the shift. A p < 0.01 threshold is required before we call a confirmed cognitive operation. p < 0.05 indicates high possibility of cognitive operation.
What makes this EU-sovereign?
Infrastructure runs on Hetzner Helsinki and Scaleway Paris. On-premises deployment is available for sovereign government clients. We are GDPR-compliant and provide data processing transparency documentation on request.
How does the training module work?
Built on Talantir, our scenario-based behavioral simulation engine, anchored in inoculation theory (McGuire 1961; van der Linden & Roozenbeek 2020). Scenarios are sourced from real DISARM-tagged retellings Retelnist observed targeting your population in the last 90 days, anonymized and ethically reviewed.
Why no self-serve demo?
Retelnist's work has operational sensitivity. We protect both our methodology and the operational privacy of current pilot partners. Every demo request is verified against organizational domain or existing referral. The 45-minute discovery call is with our methodology lead.
What DISARM version does Retelnist use?
DISARM Red Framework v2, the taxonomy adopted by EEAS, NATO StratCom CoE, and the FIMI ISAC. Tagging rules are published and auditable per-tag. LLM-assisted classification carries explicit confidence scores.
Who built Retelnist?
The methodology is co-owned by Glib Buriak, PhD, Ukrainian professor of economics and cognitive security researcher with multi-decade work on information operations doctrine. The platform is built by Andrii Zupko, founder of Talan.Tech, running parallel ventures in counter-UAS defense and behavioral simulation.
No self-serve demo. Here's why.
Request a verified briefing
Retelnist's work has operational sensitivity. We protect both our own methodology and the operational privacy of current pilot partners. Every demo request is verified.
- 01Submit organizational email and role
- 02Retelnist verifies via organizational domain or referral
- 0345-minute discovery call with our methodology lead to understand your operational environment
- 04If there's mutual fit: secured access to a live environment with pre-loaded Ukraine and Iran datasets
- 05Technical walkthrough and Q&A with your analyst team
For procurement, partnership, or methodology inquiries. PGP key available on request.
Pilot engagement
3 months
Structured pilot · one deployment
Govt & policy
On request
Annual subscription · analyst seats
NGO / Academic
Discounted
Up to 60% for verified non-commercial
Custom scenarios
Per-scope
Tailored training scenario development