AcademyOutside sources
Faculty
Methods only. Attribute by name and organization. No politics, personal beliefs, or marketing claims as fact.
A method that fails against the evidence comes out. Contested evidence is taught beside a contested method. First rule: teach the published work, not the personality. The records are on Sources.
NCI University (Neuro-Cognitive Intelligence)
Chase Hughes
Behavior profiling, elicitation, influence recognition
Taught here as method only. Marketing claims about rank or status, personal opinions, and any claim the published research does not support are left out. The deception-detection material is paired with the meta-analytic evidence on accuracy.
- Baselining before interpretation
- The six social needs and their paired fears
- Six decision-making styles
- Elicitation moves and how to notice them
Books: Six-Minute X-Ray: Rapid Behavior Profiling; The Ellipsis Manual
nci.university (opens in a new tab)
Bond and DePaulo (2006) found people average 54% correct lie-truth judgments. Behavior indicates stress or a change of state; it does not establish truth.
Complexly
Crash Course
Media literacy; navigating digital information
Video series used as supporting material for Academy lessons.
- Lateral reading
- Evaluating evidence, photos, video, and data
- Advertising, ownership, and policy
Digital Inquiry Group · Civic Online Reasoning
Digital Inquiry Group
Civic online reasoning; professional fact-checking behavior
Research-based curriculum used as supporting material.
- Lateral reading as practiced by fact-checkers
- Click restraint
CIA Center for the Study of Intelligence
Richards J. Heuer Jr.
Structured analytic techniques
Taught as published method: a reasoning discipline any investigator can adopt. Nothing about the institution that published the book is endorsed or defended here.
- Analysis of Competing Hypotheses
- Scoring evidence by inconsistency rather than by fit
- Questioning key assumptions
Books: Psychology of Intelligence Analysis (1999)
College of Policing (England and Wales); Ronald P. Fisher and R. Edward Geiselman
PEACE Model and the Cognitive Interview
Investigative interviewing
Taught as published method. PEACE is the national framework for investigative interviewing in England and Wales, and the Cognitive Interview is the memory method developed by Fisher and Geiselman. The site teaches both at a civilian scale and does not claim police authority.
- The PEACE interview structure
- The uninterrupted first account
- Cognitive Interview recall aids: context, report everything, reverse order, another point of view
Open Worldwide Application Security Project (OWASP)
OWASP Gen AI Security Project
Risks in applications built on language models
Used as the standard reference list of failure risks. The list is revised as the field changes, so the site teaches the categories rather than their ranking.
- The Top 10 for LLM Applications, including prompt injection and excessive agency
UN OHCHR and the Human Rights Center, UC Berkeley School of Law
Berkeley Protocol
Digital open source investigation: lifecycle, preservation, verification, ethics
Used as the methodological standard for the collection and verification material. The Protocol is written for investigations into serious violations of international law; this site teaches its method at a civilian scale and does not claim its authority.
- The investigation lifecycle
- Preservation before analysis
- Data minimization
- Ethical principles and investigator well-being
Books: Berkeley Protocol on Digital Open Source Investigations (2022)
www.ohchr.org (opens in a new tab)
Following a professional protocol does not make an investigator an accredited one, and does not make findings admissible anywhere.
Stichting Bellingcat
Bellingcat
Open source verification, geolocation, chronolocation, published methodology
Taught as published method: the techniques and the editorial standard, including the requirement that outside readers can reproduce the work. Specific investigations are referenced only as demonstrations of technique.
- Geolocation by convergence of fixed features
- Shadow-based time estimation
- Publishing the method alongside the finding
www.bellingcat.com (opens in a new tab)
Reproducible method is the claim being taught. Any particular investigation's conclusions are that organization's, not this site's.
US National Institute of Standards and Technology; Internet Engineering Task Force
NIST and the IETF
Digital forensics process, evidence collection, incident handling
Standards documents used directly. They are guidance for practitioners, not law, and they are revised: the incident-handling guide was substantially restructured in 2025.
- The four-phase forensic process
- Order of volatility
- Chain of custody and integrity verification
Books: NIST SP 800-86 (2006); RFC 3227 (2002); NIST SP 800-61
csrc.nist.gov (opens in a new tab)
A standard tells you the accepted process. It does not tell you what a court in your jurisdiction will accept.
US Office of the Director of National Intelligence
Analytic Standards
Analytic tradecraft standards, sourcing, expressions of likelihood and confidence
The standards are taught as a writing and reasoning discipline that any investigator can adopt. Nothing about the institution that publishes them is endorsed or defended here.
- The nine analytic tradecraft standards
- The estimative likelihood lexicon
- Separating likelihood from confidence
- Sourcing requirements
Books: ICD 203, Analytic Standards (2015); ICD 206, Sourcing Requirements
archive.dni.gov (opens in a new tab)
An institution can hold good standards and still fail to follow them. The 2005 review of the pre-war Iraq assessments is taught in this section for that reason.
Surveillance Self-Defense
Electronic Frontier Foundation
Practical digital security by threat model
Used as the how-to reference for this section. We teach the judgment and send readers there for steps, because steps go stale within months and theirs are maintained.
- Threat modeling before tools
- Step-by-step guidance we link to rather than duplicate
- The distinction between risk from a market and risk from a person
Books: Surveillance Self-Defense
ssd.eff.org (opens in a new tab)
A guide cannot know your situation. Threat modeling is the part you have to do yourself, and getting it wrong in either direction has a cost.
CyberTipline · Take It Down
National Center for Missing & Exploited Children
Child exploitation reporting and image removal
Referenced as the reporting route and the operator of the takedown service. Their published caveats are carried here rather than smoothed over.
- How the hash-based takedown services work and what they cannot do
- Where a report actually goes
www.missingkids.org (opens in a new tab)
Reporting volumes are affected by platform counting changes, and the organization says so. Trends are readable; precise year-to-year comparisons are not.
United States Secret Service
National Threat Assessment Center
Behavioral threat assessment and management; the pathway to violence
Used as published method. The material concerns identifying and intervening with a specific, already-identified person of concern whose behavior is observed over time.
- The pathway model applied to a known person of concern
- Behavioral threat assessment as an ongoing process, not a snapshot
- The distinction between threat assessment and stranger profiling
Books: Preventing Targeted Workplace Violence (Secret Service NTAC); Behavioral Threat Assessment Units: A Guide for State and Local Law Enforcement (2024)
www.secretservice.gov (opens in a new tab)
It is not a tool for assessing strangers, and applying it that way is the most common misuse of threat-assessment material.
American Association for Emergency Psychiatry
Project Beta De-escalation Workgroup
Verbal de-escalation of the agitated person
Taught as the published consensus domains, adapted in vocabulary but not in substance.
- Personal space and non-provocation
- One person establishing verbal contact
- Concise repeated language
- Naming wants and feelings
- Limits and choices
Books: Verbal De-escalation of the Agitated Patient: Consensus Statement (Western Journal of Emergency Medicine, 2012)
www.aaep.com (opens in a new tab)
It is an expert consensus statement, not a controlled trial, and it was developed for clinicians in emergency settings. The strongest experimental evidence in this area measured modest effects in trained police officers, not in ordinary people in spontaneous confrontations.
Human factors research; SA Technologies
Mica R. Endsley
Situation awareness: perception, comprehension, projection
Used as the evidence-based replacement for the color-code folklore that dominates personal-safety instruction. The model is measurable and has decades of applied validation in aviation and other high-workload domains.
- The three-level model of situation awareness
- Awareness as a built understanding rather than a readiness state
Books: Toward a Theory of Situation Awareness in Dynamic Systems (Human Factors, 1995)
journals.sagepub.com (opens in a new tab)
It describes how understanding is constructed. It is not a readiness scale and it does not claim anyone can perceive everything.
US CPSC; National Fire Protection Association
Consumer Product Safety Commission and NFPA
Carbon monoxide fatalities, generator safety, home fire causes
Federal injury-surveillance and fire-statistics data used directly for the outage wing, where the difference between folklore and data is measured in deaths.
- Portable generator carbon monoxide death figures and the 20-foot guidance
- The CO shutoff standards and their market penetration
- Home candle fire estimates
Books: CPSC, Non-Fire Carbon Monoxide Deaths Associated with the Use of Consumer Products; NFPA, Home Candle Fires
www.cpsc.gov (opens in a new tab)
Reporting periods differ between CPSC releases and should not be blended. The proposed federal generator rule was not yet final; verify its status before relying on it.
Bennett, Holloway and Farrington; Welsh and Farrington
Campbell Collaboration Reviews
Systematic reviews of neighborhood watch, CCTV and street lighting
Used because systematic reviews report their own weaknesses, which is what makes them teachable. The watch review's publication-bias finding is taught alongside its headline effect size.
- The neighborhood watch effect size and its publication bias
- CCTV effects by setting
- The lighting mechanism finding
Books: The Effectiveness of Neighborhood Watch (Campbell Systematic Reviews, 2008); Effects of Closed Circuit Television Surveillance on Crime (2008); Effects of Improved Street Lighting on Crime (2008)
onlinelibrary.wiley.com (opens in a new tab)
These reviews summarize the studies that existed. Where underlying study quality was weak, the review says so, and this site repeats it rather than quoting the headline alone.