Fire & Arson Investigator Journal

Artificial Intelligence in Fire and Explosion Investigation – A Cautionary Tale

--Manuscript Draft--

Manuscript Number:

Full Title: Artificial Intelligence in Fire and Explosion Investigation – A Cautionary Tale

Article Type: Article

Keywords: Artificial Intelligence; NFPA 921; AI; ML; template; Predictive Modeling, Fire Pattern
Analysis; Machine Learning

Corresponding Author: Joseph M Ellington, B.S.

International Association of Fire Investigators
Bowie, Maryland UNITED STATES

Corresponding Author Secondary

Information:

Corresponding Author's Institution: International Association of Fire Investigators

Corresponding Author's Secondary

Institution:

First Author: Joseph M Ellington, B.S.

First Author Secondary Information:

Order of Authors: Joseph M Ellington, B.S.

Joseph M. Ellington

Order of Authors Secondary Information:

Abstract: Artificial Intelligence (AI) is currently integrated into various technologies used by fire
and explosion investigators, including smoke and heat detector devices, fire protection
systems, cameras, and instrumentation. AI-powered analytics can generate reports
that provide valuable insights about trends or patterns in fire and explosion
investigations.
The integration of artificial intelligence in fire and explosion investigation holds
immense promise for enhanced efficiency, accuracy, and safety. However, ethical
considerations and potential biases in data must be carefully addressed. Fire
investigators must understand the algorithms they deploy and ensure the reliability and
interpretability of AI-generated results.

Additional Information:

Question Response

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
BLIND Manuscript without contact info Click here to access/download;BLIND Manuscript without
contact info;AI and Fire & Explosion Investigation.docx

1
2
3
4 ARTIFICIAL INTELLIGENCE IN FIRE AND EXPLOSION INVESTIGATION – A
5 CAUTIONARY TALE
6
7
8
9
10
11
12 ABSTRACT
13
14
15 Understanding what Artificial Intelligence (AI) is and how it evolves and potentially impacts our
16 methodology, opinions, and conclusions is essential and intriguing. It offers new possibilities and enhances
17 our capabilities, instilling a sense of optimism for the future of our field. This article aims not to teach the
18 principles of Artificial Intelligence but to explain what artificial intelligence is, increase awareness that it
19 currently exists and is applied daily in our duties and responsibilities as fire and explosion investigators,
20 and advocate caution in its development and application.
21
22 Artificial Intelligence is more than a mere buzzword; it is a pragmatic tool that is transforming the
23
24
technology employed by fire and explosion investigators. Its real-world applications are reshaping our
25 investigative approach, presenting us with fresh opportunities, and amplifying our capabilities.
26
27 While Artificial Intelligence has undoubtedly emerged as a potent tool for addressing complex problems, it
28 is not without its perils. In this article, we delve into the role of AI in boosting investigation efficiency and
29 the advantages it offers to professionals in the field. This paper clarifies what artificial intelligence is and
30 how it is integrated into various technologies, including those used by fire and explosion investigators. It
31 underscores the imperative of a measured approach that harnesses artificial intelligence's potential power
32
while underscoring the crucial need to exercise caution in its development and application, fostering a sense
33
34 of responsibility in utilizing this formidable tool.
35 NFPA 921 - Guide for Fire and Explosive Investigations is considered the standard of care in fire
36
37
investigation and is widely used by fire investigators. It is also a critical resource that can inform the
38 integration of emerging technologies and explain the crucial interplay with NFPA 1033. We must
39 continuously update our knowledge regarding the content and changes to these standards. The lack of such
40 knowledge may render an expert 'unqualified' to testify in a legal proceeding. Neither NFPA 921 nor NFPA
41 1033 acknowledges or directly addresses artificial intelligence, highlighting the need for us to stay informed
42 and adapt to the changing landscape of our field.
43
44
AI-powered analytics can generate reports that provide valuable insights about trends or patterns in fire and
45
46 explosion investigations. This enhances our understanding of these incidents and streamlines the reporting
47 process, making it more efficient. In report writing, limited aspects of AI are already used to research
48 definitions and issues related to the origin, cause, and responsibility for fires and explosions. It can also ask
49 questions from documents and create summaries of depositions used for legal proceedings, ensuring that
50 our legal processes are thorough and efficient. While AI can undoubtedly improve the efficiency and speed
51 of report writing, it has limitations. It does not meet existing legal requirements regarding its content and
52 identifying sources of information.
53
54
55
56 INTRODUCTION
57
58 A cautionary tale is told in folklore and ubiquitous in popular culture that warns listeners of dangers and
59 spreads awareness. Many urban legends are framed as cautionary tales, from The Lover’s Lane, haunted
60
61
62
63
64
65
 1
2
3
4 by a hook-handed murderer, to the story of a man who shot a cactus for fun only to die when the plant
5 toppled onto him.1
6
7
8 A cautionary tale often relies on analogy, a comparison of two things for explanation or clarification, or
9 metaphor, a figure of speech or idea used in place of another to suggest an analogy or similarity between
10 them. In this paper, the ‘woods’ in Little Red Riding Hood, for example, is an analogy or metaphor for
11 artificial intelligence. Unlike most cautionary tales, this paper, in addition to simply warning of the potential
12 dangers inherent in artificial intelligence, explains how AI is currently embedded in the present fabric of
13 our discipline and offers solutions for bettering its future application.
14
15
16
17
18 Countless versions of “Little Red Riding Hood” exist, many of
19 which reflect the concerns and moral issues of the period in
20 which each version first appeared. The story originated in
21 the Middle Ages in Europe as a cautionary tale to warn small
22
23
children about the dangers of the woods. (“Little Red Riding
24 Hood | Story, Characters, Grandma, Moral, & Themes”)
25
26
27
28
29
30
31 Figure 1 – Little Red Riding Hood. An example of a Cautionary Tale
32
33 While NFPA 921- Guide for Fire and Explosion Investigations does not directly address Artificial
34 intelligence, the document remains a critical resource for fire investigators. Its principles can inform the
35 integration of emerging technologies in the field. At the very least, fire and explosion investigators must
36 understand what artificial intelligence is, where and how it currently permeates our discipline, and to what
37 degree it will impact our profession. This understanding will empower us to adapt and utilize AI effectively
38 and positively in our investigations.
39
40
41 The National Fire Protection Association created a set of investigative guidelines in 1992 that was published
42 as “NFPA 921: Guide for Fire and Explosion Investigations.” Both public and private fire investigators
43 widely use the guide, which describes in detail the scientific method and how to apply it in fire and
44 explosion investigations. The stated purpose of the guide has been and continues to be to “improve the
45 probability of reaching sound decisions.” 2 The reader of this article and NFPA 921 should realize that such
46 an outcome is never guaranteed, only that the probability of its outcome is increased. The bedrock of
47 science is not its certainty but uncertainty. We can only judge how accurate or correct something is by
48
identifying and stating its probability.
49
50
51 For this reason, NFPA921 requires an investigator to know the certainty needed to express an expert
52 opinion. Only when the opinion is considered ‘probable’ should it be described with reasonable certainty.
53 Otherwise, anything is possible if the certainty level is only suspected.
54
55 Over time, NFPA 921’s recommendations have grown to include guidelines for determining the origin and
56 cause of fires, collecting and managing physical evidence, and chapters focusing on motor vehicle fires,
57
58
59 1 Cautionary Tale – From Wikipedia, the free library
60 2 2024 - NFPA 921 – Guide for Fire and Explosion Investigations – Chapter 1.2 Purpose – Section 1.3
61
62
63
64
65
 1
2
3
4 wildfires, and appliances. The standard was recently changed in the 2024 edition to enhance its
5 effectiveness and address evolving needs in fire and explosion investigations. Specifically, the 2024 edition
6
7
provides additional guidance on bias and expressions of certainty on conclusions, the impact of suppression
8 activities, fire effects on electrical systems, and updates concerning fire patterns, arc mapping, and fire
9 classification. These updates aim to improve the accuracy, reliability, and consistency of investigations and
10 include the following:
11
12 1. Advancements in Science and Technology - As our understanding of fire dynamics, materials, and
13 forensic techniques evolves, the guide must incorporate the latest scientific and technological
14 advancements.
15
2. Feedback from Practitioners – Input from fire investigators, legal professionals, and other
16
17 stakeholders is crucial.
18 3. Legal and insurance requirements - The standard is a reference for legal proceedings and insurance
19 claims. Its updates strive to align with changing legal requirements and industry practices.
20 4. Improved Clarity and Consistency - The 2024 edition clarifies concepts, provides additional
21 guidance, and promotes investigation consistency, helping investigators reach accurate
22 conclusions.
23
24 Whether adopted into law or not, NFPA 921 is regarded as the standard of care in fire investigation.
25
26
Technically, its provisions are still guidelines, but realistically, they are considered authoritative and
27 utilized by many courts when determining the sufficiency and admissibility of expert opinions. Compliance
28 with these guidelines is necessary to prepare and accept an admissible expert opinion.
29
30 The admissibility of an expert opinion also requires that the expert be qualified. The National Fire
31 Protection Association provides its standards for the minimum professional qualifications for a fire
32 investigator in NFPA 1033—Standard for Professional Qualifications for Fire Investigators. The standards
33 require an expert to maintain updated basic knowledge in several areas, including, but not limited to, fire
34
chemistry and thermodynamics. Additionally, NFPA 1033 requires updated knowledge regarding fire
35
36 investigation methodology and evidence documentation.
37 This requirement creates a critical interplay between NFPA 921 and NFPA 1033. Not only is it beneficial
38
to have an expert who understands the current NFPA 921 guidelines utilized by most courts, but a failure
39
40 to be aware of any changes or updates to NFPA 921 arguably renders an expert “unqualified.” A lack of
41 knowledge regarding the content and changes to NFPA 921 is the equivalent of a failure to satisfy the NFPA
42 1033 requirement that an investigator stays up to date on investigation methodology.
43
44 Neither NFPA 921 nor NFPA 1033 includes the term artificial intelligence. Despite this, both standards
45 focus on traditional investigative methodologies and do not explicitly delve into artificial
46 intelligence or machine learning techniques. Indirectly, each recognizes the potential impact of AI (artificial
47 intelligence) and ML (machine learning) on enhancing investigative practices.
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1
2
3
4 At its most superficial level, Artificial intelligence (AI) refers to a computer or computer-controlled robot's
5 ability to perform complex tasks that historically only
6
7
humans could do, such as reasoning, making decisions,
8 or solving problems commonly associated with
9 intelligent beings.
10
11
Today, “AI” describes a wide range of technologies
12
13 that power many of the services and goods we use
14 daily, from apps that recommend TV shows to chatbots
15 that provide customer support in real-time.
16
17 Figure 2 - AI systems can process enormous amounts
18 of data, identify patterns, and make decisions based
19 on collected information.
20
21
22 ARTIFICIAL INTELLIGENCE EXAMPLES
23
24 Artificial intelligence (AI) is the theory and development of computer systems capable of performing tasks
25
that historically required human intelligence, such as recognizing speech, making decisions, and identifying
26
27 patterns—AI technologies include machine learning, deep learning, and natural language processing (NLP).
28
29 "Though the humanoid robots often associated with Aias seen in the contemporary I, ROBOT, film doesn’t
30 exist yet; you’ve likely interacted with machine learning-powered services or devices many times before." At
31
32
the most superficial level, machine learning uses algorithms trained on data sets to create machine learning
33 models that allow computer systems to perform tasks like making song recommendations, identifying the
34 fastest way to travel to a destination, or translating text from one language to another.
35
36
ARTIFICIAL INTELLIGENCE IN THE FIRE SERVICE – THE PRESENT
37
38
39 "AI has transformed almost all professions by developing advanced tools and techniques. Some of the
40 most common examples of AI in use today include:
41
42
 ChatGPT: Uses large language models (LLMs) to generate text responding to questions or
43
44 comments.
45  "Google Translate: Uses deep learning algorithms to translate text from one language to another."
46  Netflix: Uses machine learning algorithms to create personalized recommendation engines for users
47 based on their previous viewing history.
48  "Tesla: Uses computer vision to power self-driving features on their cars.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1
2
3
4 The integration of AI into the fire service brings about significant efficiency improvements. AI streamlines
5 time-consuming tasks like data collection, analysis, and report generation through automation. By
6
7
automating these processes, users can focus on
8 higher-value tasks. In the fire service, artificial
9 intelligence can help respond to and assess a fire
10 scene to make decisions about extinguishing a
11 fire and limiting its spread and damage.
12
13
14
15
16
17
18
19
20 Figure 3 – Using Artificial Intelligence in the Fire Service
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37 Figure 4 – Collected data can be used in real-time to suppress or extinguish a fire.
38
39 AI revolutionizes firefighting by providing critical insights, improving safety, and enhancing response
40 capabilities. It is currently used in flashover prediction, data interpretation, on-scene data collection, and
41 real-time traffic analytics. Artificial intelligence data can also be used to assess building and fire conditions.
42
43
44 In fire and explosion investigation, artificial intelligence may be used for hypothesis generation, decision-
45 making, and strategy development. AI's ability to manage repetitive and mundane tasks allows investigators
46 to dedicate more time to critical thinking and problem-solving, resulting in faster and more accurate
47 investigation outcomes.
48
49 Machine learning and natural language processing empower investigators to analyze vast quantities of
50
digital evidence rapidly and accurately. By identifying patterns, anomalies, and connections, AI algorithms
51
52 uncover hidden insights that might otherwise remain concealed. These AI-powered investigation techniques
53 optimize time and boost efficiency and effectiveness.
54
55 As an example, an investigator or expert finds a restaurant extinguishing system did not function correctly
56 during a fire, either failing to put out a fire or allowing it to spread, and observes crystallization in the liquid
57 extinguishing agent during a laboratory examination after opening the extinguishing canister. A search
58
59
60
61
62
63
64
65
 1
2
3
4 using COPILOT 3 (an application based on artificial intelligence) returns the following possible reasons to
5 explore (e.g., changes in temperature, pressure, or chemical composition.) The findings would point the
6
7
investigator to evaluate or explore whether any conclusions were consistent with the facts and data before
8 developing a final hypothesis concerning the fire's origin, cause, and responsibility.
9
10 COPILOT results for the question “What are the reasons for crystallization in ANSUL systems?” -
11 ANSUL systems are industry-standard fire suppression systems used in environments like
12 restaurants. They work by discharging a mixture of wet chemicals when fires are detected, smothering the
13 flames and preventing them from reacting with oxygen. (“Maintaining Your ANSUL System in A
14
Restaurant: What You Need to Know”)
15
16 However, the search results do not explicitly mention the reasons for crystallization in ANSUL systems.
17
Crystallization could occur due to a range of factors, such as changes in temperature, pressure, or
18
19 chemical composition, but without specific information, it is hard to determine the exact cause.
20
21 Proper maintenance of these systems is essential to ensuring their effectiveness. This includes regular
22 professional inspections, good restaurant practices to prevent grease build-up, and staying current on
23 documentation. If you are experiencing issues with crystallization in your ANSUL system, it would be
24 best to consult a professional or the manufacturer for guidance.
25
26
27
28
Fire and explosion investigation is a complex process involving knowledge spanning several disciplines.
29 While a report may be prepared for any number of reasons, its primary purpose is to identify and describe
30 where a fire or explosion began (i.e., its origin), what caused it (i.e., its cause), and responsibility (who or
31 what was accountable for its damage.) You may have encountered examples of it without knowing.
32
33 Artificial Intelligence (AI) is currently incorporated into technology used by fire and explosion investigators
34 in the following ways:
35
36
37 Smoke and heat detector devices and components of fire protection systems: Artificial Intelligence has
38 transformed investigations by developing advanced tools and techniques. It plays a crucial role in analyzing
39 data from smoke and heat detectors, cameras, and instrumentation and processing data from equipment
40 used to detect and analyze debris and artifacts. AI is used to analyze data to design, place, and collect data
41 from these devices to detect smoke and flames, identify patterns, and make decisions about the presence of
42 fire or smoke.
43
44
45 AI records, captures, and measures fire and explosion events with cameras and video. It can analyze the
46 data collected by these devices to identify essential information related to the origin and cause of the
47 incident. Equipment used to detect and analyze debris and artifacts. AI is incorporated into equipment that
48 processes data from hydrocarbon detectors, CT scans, GCMS, and MS (Gas Chromatography-Mass
49 Spectrometry) devices. This helps investigators identify specific substances or chemicals at the scene and
50 determine their relevance to the fire or explosion.
51
52
53 Digital evidence analysis is crucial in forensic investigations, encompassing various data sources like
54 emails, social media, financial transactions, and surveillance footage. AI algorithms are instrumental in
55 analyzing and interpreting this vast data, revealing relevant patterns and information. Machine learning
56
57 3
COPILOT - Microsoft developed COPILOT, part of the Windows operating system. It is designed to assist users in
58 providing information, answering questions, and engaging in conversation. It uses advanced language models and
59 other technologies to understand and communicate fluently. It can help with various tasks, from generating content
60 to providing information on complex topics.
61
62
63
64
65
 1
2
3
4 algorithms excel at identifying similarities, outliers, and suspicious activities, enabling investigators to
5 connect the dots and build strong cases. By leveraging AI-powered forensic tools, investigators can
6
7
efficiently manage the growing complexity and volume of digital evidence.
8
9 AI powers many search engines used by fire and explosion investigators. These AI-powered search engines
10 can process enormous amounts of data and provide relevant information related to fire investigation
11 techniques, legal requirements, and other relevant topics. “AI-powered analytics already drill down into
12 databases and documents and auto-produce easy-to-understand written reports that unearth insights about
13 trends or patterns that otherwise may never have seen the light of day.” [4] AI analytics can generate reports
14
that provide insights about trends or patterns in fire and explosion investigations.
15
16
17 Ultimately, fire investigators and other experts author, review, and publish origin and cause reports for their
18 clients and the legal system, some of which opine on responsibility. Limited aspects of artificial intelligence
19 are already used in the fire and explosion sector to generate reports about the origin, cause, and
20 responsibility for fires and explosions. This author has used AI to research definitions and issues related to
21 fire, explosions, and litigation, to ask questions from documents, and to create summaries of depositions on
22 behalf of clients.
23
24
25 Artificial Intelligence has already entered the current report-writing process on some levels. It will continue
26 to evolve into other aspects of fire and explosion investigation and litigation in the future. While AI can
27 improve the efficiency and speed of report writing, it falls short of meeting existing legal requirements
28 regarding its content and identifying sources of information.
29
30 Every report should be considered an expert report. A witness who is qualified as an expert by knowledge,
31
skill, experience, training, or education may testify in the form of an opinion conclusion or otherwise if the
32
33 expert’s scientific, technical, or other specialized knowledge will help the trier of fact (e.g., judge or jury)
34 understand the evidence or determine a fact in issue.
35
36 A report is only the visible portion of the iceberg in fire and explosion investigations. The fire investigator’s
37 report is the most critical documentary evidence. Despite this, less attention is paid to reports than to an
38 investigation by the author and the intended audience. It is vital that such testimony is the product of reliable
39 properties and methods and that the expert reliably applies verifiable principles and techniques to the facts
40 of the case.
41
42
43 To what degree artificial intelligence affects the substance of such reports, and not just its appearance or
44 form, is subject to question and discussion. Grammarly, for example, incorporates AI and claims to
45 scrutinize your writing to improve clarity and word choice. It also offers free and fee-based services. The
46 selection of words and use of language affects not only the appearance of the expert's report but also its
47 logic. Grammarly goes beyond checking spelling and grammar but also affects clarity, engagement, and
48 delivery. Any attorney will tell you how you say something, orally or in writing, affects its meaning. AI
49
can improve the efficiency and accuracy of investigations, but it does not replace the need for human
50
51 expertise and input. What is certain is that it and other tools need human input for analysis and narrative.
52
53 Authoring reports based on the analysis of a documented investigation involves time and money in the real
54 world. Fire and explosion investigations and documentation are lengthy and complex, and the specific
55 processes resulting from a report are not necessarily efficient or cost-effective. Accordingly, much effort
56 has been invested in developing ways and means to investigate a scene and generate reporting more
57 efficiently.
58
59
60
61
62
63
64
65
 1
2
3
4 Currently, templates are the most popular and common method by which organizations classically approach
5 the generation of reports. This method typically involves an organization-wide acceptance of a specific
6
7
format with internal references that comply with external protocols and standards (e.g., NFPA 921, NFPA
8 1033, ASTM standards). The approach is implemented through internal technical and administrative
9 reviews. It is a quality control approach to ensure the content and format of reports through review before
10 the report is published or released.
11
12 The template technique may be replaced or supplemented with an ‘active document’ approach that
13 automates the process by combining data extracted automatically, instantaneously, and from and remotely
14 from the scene with the accepted internal format to produce a finished report. Organizations may also
15
choose to reach out to other parties who specialize in developing reports based on these approaches and are
16
17 attempting to balance internal (e.g., organizational) and external (e.g., clients or the legal system) needs and
18 demands.
19
20 Organizations may also adopt the generation of reports through external parties and incorporate them with
21 their branding. Many of these third parties advertise that the finished report may be generated either from
22 live data recorded by the investigator from the scene or
23 generated from their notes and will meet the requirements
24 of NFPA 921 and NFPA 1033.
25
26
27 At best, using a template guarantees uniformity
28 throughout the organization and saves money and time.
29 However, it does not guarantee that the investigator's
30 opinions and conclusions regarding the facts and
31 evidence (hypothesis) are correct.
32
33 Figure 5 – Origin and Cause templates based on artificial intelligence.
34
35
36 In the future, expect Artificial Intelligence in many reports regarding fires or explosions you receive,
37 certainly in form or appearance and progressively in substance (e.g., opinions and conclusions). Likely,
38 artificial intelligence may eventually replace technical and administrative review of reports if for no other
39 reason than it already has, or will become, more efficient and cost-effective.
40
41 While AI solves many of the difficult choices to be made in establishing where a particular methodology
42 (e.g., NFPA-921 Guide for Fire and Explosion Investigations) or requirement (e.g., NFPA 1033 Standard
43 for Fire Investigator Qualifications) has been consulted and adhered to, it falls short of meeting existing
44
45
legal requirements in terms of content.
46
47 Limited aspects of AI are already used in report writing. AI can assist in researching definitions and issues
48 related to fire and explosions, asking questions from documents, and creating summaries of depositions. It
49 can also help automate the generation of reports by combining data extracted from the scene with the
50 accepted internal format of the report.
51
52 The guidelines for expert witnesses in federal court are primarily outlined in Rule 26 of the Federal Rules
53
of Civil Procedure.4, which can be downloaded from the United States Courts website at www.uscourts.gov.
54
55 While some states accept and implement the federal standard, every report should be considered an expert
56 report, regardless of whether it is in the federal system.
57
58
59 4
www.uscourts.gov
60
61
62
63
64
65
 1
2
3
4 The reader of this article would also do well to review the Federal Plain Language Guidelines promulgated
5 as law in 2010. Although these guidelines were developed to help federal agencies write documents clearly
6
7
and understandably so that users can easily find, understand, and use the information they need, they can
8 also be a valuable resource for expert witnesses who investigate and write reports concerning the origin,
9 cause, and responsibility for fire and explosions and help to understand their audience, organize the report,
10 apply sound writing principles, and test the document’s effectiveness in reaching its intended audience.
11
12 While no rule prevents opposing legal counsel from asking a witness to speculate, generally avoid doing
13 so5, the author of a report must testify under oath to its content and conclusions and identify all sources of
14 information on which they consulted or relied. Artificial Intelligence currently does not provide this
15
information. However, it is cost-effective, overcomes writer's block, creates content faster than people, and
16
17 is economical. Unfortunately, artificial intelligence frequently misrepresents or does not fully explain the
18 facts and, at least in some cases, disguises incompetence and allows bias to enter the system.
19
20 While AI solves many of the difficult choices in establishing whether a particular methodology (NFPA-
21 921) or requirement (NFPA 1033) has been consulted and adhered to, it also falls short of meeting existing
22 legal requirements regarding its content. While truth in the legal system may be determined by the vote of
23 a jury and based on their assessment of the facts and evidence presented to them, artificial Intelligence does
24 not provide this same information nor abide by guidelines, standards, or the rules of a court. It is, however,
25
26
cost-effective, overcomes writer's block, and creates content faster than people (which is the most
27 significant benefit.)
28
29 The presence of Artificial Intelligence in a report is challenging to detect, and the line that identifies its
30 presence and inclusion is becoming more blurred with each passing day. Clues that indicate the possible
31 presence of Artificial Intelligence are a lack of typos, overuse of ‘the,’ no cited sources, shorter sentences,
32 repeated words or phrases, and the lack of analysis. Problematically, the same characteristics can surface
33 simply in a report that is reviewed administratively or technically.
34
35
36 Artificial intelligence can create realistic images from textual input or information (e.g., in the photo at the
37 left) or enhance existing images. Examples of resources that can help create images are Canva’s AI Image
38 Generator, Hotspot.ai, ImagineArt AI Generator, and OpenArts’s Image Generator.
39
40 An AI-generated or enhanced image can be used as evidence in reports
41 to convince others (e.g., other experts, attorneys, or a judge and jury) to
42 interpret evidence differently. Fortunately, the GPS and meta-data
43 recorded when fire and explosion investigators create or collect images
44
45
cannot be modified or erased from their cameras' original raw data,
46 assuming it is still available for the reasonable examination of others.
47
48 For this reason, some are unwilling to provide the original data and
49 instead offer digital copies. One example is PDFs that have either
50 removed or erased essential information concerning where and when the
51 image was collected and whether it has since been modified.
52
53
Figure 6 – Image created using artificial intelligence.
54
55
56
57
58
59 5Ellington, J.M. (2024) Blog on Artificial Intelligence and Fire Investigation. – http://firelogix.net/my-
60 blogs/f/artificial-intelligence
61
62
63
64
65
 1
2
3
4 ARTIFICIAL INTELLIGENCE IN FIRE AND EXPLOSION INVESTIGATION6
5
6
7 Fire Pattern Analysis: AI algorithms are instrumental in deciphering complex fire patterns. By processing
8 images and data from fire scenes, AI systems can identify distinctive patterns that may elude the human
9 eye. This aids investigators in reconstructing the sequence of events leading to a fire.
10
11
12 Predictive Modeling: Machine Learning algorithms can be trained on historical fire data to develop
13 predictive models. These models help anticipate potential fire hazards, assess the risk of specific scenarios,
14 and enable proactive design, prevention, and investigation measures.
15
16 Evidence Analysis: AI technologies facilitate rapid and precise analysis of evidence collected from fire
17 scenes. Whether identifying accelerants, analyzing burn patterns, or categorizing fire-related artifacts, AI
18
enhances the speed and accuracy of evidence interpretation.
19
20
21 Real-time Monitoring: Integrating AI with sensors and monitoring devices allows for real-time assessment
22 of fire-prone areas. AI algorithms can analyze environmental conditions, detect anomalies, and alert
23 authorities to potential fire risks, enabling swift intervention.
24
25
26 THE FUTURE – CHALLENGES AND ETHICAL CONSIDERATIONS
27
Artificial intelligence currently includes a wide range of applications with the potential to transform how we
28
29 work and our daily lives. It is a complicated picture of potential benefits and dangers. While integrating AI
30 and ML in fire and explosion investigation in the future holds immense promise, it also presents challenges,
31 including ethical considerations. Ensuring the reliability and interpretability of AI-generated results is
32 crucial. Fire investigators must understand the algorithms they deploy and be cautious of potential biases
33
34
in data that might impact the accuracy of their predictions.
35
Ethical considerations include the responsible use of AI in investigations, preserving privacy rights, and
36
37 maintaining transparency in decision-making processes. Striking a balance between harnessing the benefits
38 of technology and upholding ethical standards is essential for the responsible adoption of AI and ML in fire
39 investigations.
40
41 As we navigate the evolving landscape of fire investigation, integrating Artificial Intelligence and Machine
42 Learning leads toward enhanced efficiency, accuracy, and a safer future. Merging innovative technology
43 and established industry standards propels us into a new era of investigative possibilities and accuracy.
44 With the power to unravel complex fire patterns, predict potential hazards, and assist in cause
45 determination, AI and ML are not merely tools but allies in the pursuit of truth. As we embrace these
46 advancements, we must move forward carefully, acknowledging the challenges and ethical considerations.
47
48
49 In fire investigation, where precision and reliability are paramount, the synergistic dance between human
50 expertise and technological innovation promises a future where mysteries are unraveled, risks are mitigated,
51 and justice is served. As we move into this future, guided by industry standards and a commitment to
52 responsible implementation, we look to a future where the fusion of human intellect and artificial
53 intelligence offers us the promise of gains in fire investigation.
54
55
56
This author believes and predicts that artificial intelligence has already found its way into the current report-
57 writing process on some levels and will continue to do so to varying degrees throughout our
58
59 6 "Unveiling the Future: Harnessing AI and Machine Learning in Fire Investigation" - Ohio Arson School
60 (ohioarsonschoolinc.org)
61
62
63
64
65
 1
2
3
4 profession. Until methods become available, and this becomes clear, fire and explosion investigators must
5 continue to know and apply accepted principles required by guidelines and standards recognized and
6
7
accepted by their discipline and the legal system. They must also understand artificial intelligence when
8 conducting investigations and reaching opinions and conclusions regarding origin, cause, and
9 responsibility.
10
11 To what degree Artificial Intelligence will evolve and permeate the entire fire and explosion investigative
12 process and eventually become reflected in NFPA Standards 921 and 1033 depends on time, technology,
13 and the legal system. Regardless, our actions, opinions, and findings must remain truthful, documentable,
14 and verifiable to be credible.
15
16 While many of these transformations offer exciting benefits, like self-driving cars, virtual assistants, or
17 wearable devices in the healthcare industry, they pose many challenges. The landscape of fire and explosion
18
19 investigation is transforming as it embraces further advancements in Artificial Intelligence (AI) and
20 Machine Learning (ML). Reality is likely to be more complex than is currently imagined. Truth always turns
21 out stranger than fiction. Ultimately, we would be better served not to shift blame but to realize that the
22 responsibility is ours to shoulder, “that doing wrong that good may come of it is never right and that we must
23 act right whatever the consequence.” 7
24
25
26
27 1. Cautionary Tale – From Wikipedia, the free library.
28
29
2. 2024 - NFPA 921 – Guide for Fire and Explosion Investigations – Chapter 1.2 Purpose – Section 1.33.
30 COPILOT - Microsoft developed COPILOT, which is part of the Windows operating system. It is
31 designed to assist users in providing information, answering questions, and engaging in conversation. It
32 uses advanced language models and other technologies to understand and communicate fluently. It can
33 help with various tasks, from generating content to providing information on complex topics
34 4. Ellington, J.M. (2024) Blog on Artificial Intelligence and Fire Investigation. – http://firelogix.net/my-
35 blogs/f/artificial-intelligence
36 5. The guidelines for expert witnesses in federal court are primarily outlined in Rule 26 of the Federal
37
Rules of Civil Procedure, which can be downloaded from the United States Courts website at
38
39 www.uscourts.gov. While some states accept and implement the federal standard, every report should be
40 considered an expert report, regardless of whether it is in the federal system.
41 6. "Unveiling the Future: Harnessing AI and Machine Learning in Fire Investigation" - Ohio Arson
42 School (ohioarsonschoolinc.org)
43 7. Plain Language Guidelines http://www.plainlanguage.gov/media/FederalPLguidelines.pdf
44 8. H.W. Crocker III, Robert E. Lee on Leadership – Executive Lessons in Character, Courage, and
45 Vision,1999, FORUM - A Prima Publishing Company
46
47
48
49
50
51
52
53
54
55
56
57
58
59 7H.W. Crocker III, Robert E. Lee on Leadership – Executive Lessons in Character, Courage, and Vision,1999,
60 FORUM - A Prima Publishing Company
61
62
63
64
65
Title Page with Bios, Abstract, Keywords

Author: Joseph M. Ellington, IAAI-CFI

Artificial Intelligence in Fire and Explosion Investigation – A Cautionary Tale
FireLogix
137 Arrowwood St.
Lake Jackson, TX 77577
832-544-4247
www.firelogix.net
Bios
Mr. Ellington is an internationally certified fire and explosion investigator with more than 40
years of experience. He has published articles and written books on advanced fire and explosion
investigation. His specific areas of experience include residential, business, commercial, and
industrial losses involving consulting in the disciplines of advanced fire and explosion dynamics,
computer modeling, and artificial intelligence Mr. Ellington has investigated the origin, cause,
and responsibility for injury and death cases from fires and explosions both in state and federal
courts.

Abstract: Artificial Intelligence (AI) is currently integrated into various technologies used by fire
and explosion investigators, including smoke and heat detector devices, fire protection systems,
cameras, and instrumentation. AI-powered analytics can generate reports that provide valuable
insights about trends or patterns in fire and explosion investigations.

The integration of artificial intelligence in fire and explosion investigation holds immense promise
for enhanced efficiency, accuracy, and safety. However, ethical considerations and potential
biases in data must be carefully addressed. Fire investigators must understand the algorithms they
deploy and ensure the reliability and interpretability of AI-generated results.

Keywords:

Artificial Intelligence; template; Predictive Modeling, Fire Pattern Analysis; Machine Learning
Figure 1 Click here to access/download;Figure;Figure 1 Little Red Riding Hood.png
Figure 2 Click here to access/download;Figure;Figure 2 Artificial Intelligence.png
Figure 3 Click here to access/download;Figure;Figure 3 AI in the Fire Service.png
Figure 4 Click here to access/download;Figure;Figure 4 Artificial Intelligence in the Fire
Service.png
Figure 5 Click here to access/download;Figure;Figure 5 - Template Reports.png
Figure 6 Click here to access/download;Figure;Figure 6 AI Image.png