Loading
Expert Witness Insights by Area of Expertise
13 min read
Areas of Expertise :
Civil Engineering, Artificial Intelligence, Digital Forensics, Drones, 3D Printing
Published Date:
Last Modified Date:
Emerging civil engineering technologies can create legal and evidentiary risks involving system reliability, professional responsibility, data preservation, and allocation of project duties. This article examines AI, digital twins, BIM, drones, 3D-printed construction, and embedded sensors, including the digital evidence attorneys should preserve and the methodology issues that may limit or exclude expert testimony.
Civil engineers use artificial intelligence, digital twins, drones, building information modeling, 3D-printed construction, and embedded sensors to help design, inspect, build, and monitor infrastructure.
These tools can improve how engineers collect information, evaluate conditions, and coordinate project work. They can also create difficult questions when data are incomplete, a system produces an unreliable result, or project documents do not clearly assign responsibility for reviewing, validating, approving, or acting on the output.
This guide explains several legal and evidentiary risks attorneys may need to investigate, the digital information they should consider preserving, and the methodology problems that can weaken or exclude expert testimony.
When evaluating a technology-related failure, counsel will often need to determine who controlled the process, who reviewed or relied on the result, and whether any party had a duty to identify, report, correct, or respond to the problem.
Depending on the claims and project structure, counsel may need to examine the roles of engineers, contractors, owners, software vendors, drone operators, and technical consultants. Relevant questions may include who selected the technology, supplied the data, configured the system, approved its use, controlled updates, or agreed to review its output.
An engineer’s selection, use, review, or reliance on software may become relevant when a party alleges that the engineer’s work fell below the applicable professional standard of care. Claims involving a technology vendor will depend on the causes of action asserted, the governing law, the relationship between the parties, and the terms and enforceability of any warranties, disclaimers, indemnity provisions, or liability limitations.
The specific claims and duties will depend on the governing law, contract terms, and facts of the project.
To reduce uncertainty, project teams may consider addressing:
A contract that simply requires a party to “use AI” or “provide a BIM model” may leave important questions about reliance and responsibility unresolved.
Construction Engineering
Digital Modeling
Unmanned aerial vehicle uav
Aerial Mapping
Light detection and ranging lidar
3D Printed Homes
Sensors
Sensor Processing
Digital Evidence
Data Analysis
Civil engineers use artificial intelligence, digital twins, drones, building information modeling, 3D-printed construction, and embedded sensors to help design, inspect, build, and monitor infrastructure.
These tools can improve how engineers collect information, evaluate conditions, and coordinate project work. They can also create difficult questions when data are incomplete, a system produces an unreliable result, or project documents do not clearly assign responsibility for reviewing, validating, approving, or acting on the output.
This guide explains several legal and evidentiary risks attorneys may need to investigate, the digital information they should consider preserving, and the methodology problems that can weaken or exclude expert testimony.
When evaluating a technology-related failure, counsel will often need to determine who controlled the process, who reviewed or relied on the result, and whether any party had a duty to identify, report, correct, or respond to the problem.
Depending on the claims and project structure, counsel may need to examine the roles of engineers, contractors, owners, software vendors, drone operators, and technical consultants. Relevant questions may include who selected the technology, supplied the data, configured the system, approved its use, controlled updates, or agreed to review its output.
An engineer’s selection, use, review, or reliance on software may become relevant when a party alleges that the engineer’s work fell below the applicable professional standard of care. Claims involving a technology vendor will depend on the causes of action asserted, the governing law, the relationship between the parties, and the terms and enforceability of any warranties, disclaimers, indemnity provisions, or liability limitations.
The specific claims and duties will depend on the governing law, contract terms, and facts of the project.
To reduce uncertainty, project teams may consider addressing:
A contract that simply requires a party to “use AI” or “provide a BIM model” may leave important questions about reliance and responsibility unresolved.
AI can support a range of civil engineering tasks, including data analysis, project planning, monitoring, and risk assessment. Reliability problems can arise when users accept an AI output without checking the underlying data, the system’s intended use and limitations, or the need for professional review.
ASCE Policy Statement 573 states that civil engineers must maintain professional responsibility when AI is used in civil engineering work and that AI cannot replace the professional judgment of a licensed professional engineer.
The policy provides professional guidance rather than a uniform rule of law. Depending on the jurisdiction and the evidence presented, it may still be considered when evaluating professional practices or the applicable standard of care.
An expert who merely repeats an AI-generated conclusion without establishing a reliable foundation may have difficulty satisfying Federal Rule of Evidence 702. That foundation may include the underlying data, validation, system version, intended use, and relevant human oversight.
One potential source of disagreement is version control, especially when project participants rely on different versions of the same model.
Building information modeling is a process for creating, managing, and coordinating digital information about a facility across design, construction, and operations. A digital twin may build on a BIM model by incorporating sensor data, operational information, and current or historical records associated with the physical asset.
Reliability questions may involve outdated versions, missing updates, incomplete sensor data, restricted access to source information, or use beyond the model’s intended purpose.
Detail is not the same as accuracy. An expert should determine whether the model reflects the relevant condition and date.
For example, if an expert cannot identify which model version produced a visualization or explain how missing sensor data were handled, the analysis may be difficult to authenticate, test, reproduce, or defend under Rule 702. Authentication of the item and reliability of the expert’s opinion remain separate legal questions.
Drones are used for surveying, inspections, mapping, documentation, and other technical investigations. A polished image or three-dimensional model may still raise questions about authentication, accuracy, reliability, scale, geolocation, or completeness if the collection or processing steps are poorly documented.
Many nonrecreational civil operations involving small unmanned aircraft systems are conducted under 14 C.F.R. Part 107. Other statutory, regulatory, exemption, waiver, or public-aircraft pathways may apply depending on the aircraft and operation.
Depending on the technical issues in the case, counsel may consider preserving:
Part 107 does not require every item on this litigation-preservation list in every operation. Depending on the dispute, these materials may still help counsel evaluate authentication, reliability, accuracy, and reproducibility.
Compliance with Part 107 also does not, by itself, establish that the resulting imagery or analysis is accurate or admissible.
3D-printed construction raises questions about material consistency, layer bonding, curing, reinforcement, tolerances, testing, and applicable codes or approval requirements.
An investigation may need to examine the printing equipment, material composition or mixture, environmental conditions, digital design files, toolpath instructions, printing sequence, interruption history, reinforcement placement, and available quality-control information.
Because the structure is created layer by layer, problems with material consistency, bonding, or the printing process may weaken structural performance.
Embedded sensors can collect data about structural behavior and surrounding conditions. Conclusions based on sensor data may become less reliable when the data are incomplete, collection conditions are poorly documented, sensors are not properly calibrated, or the analytical model does not reflect the structure.
Depending on the issue, an expert may examine how the data were collected, whether the sensors were calibrated and functioning properly, which conditions were measured, how the data were processed, and whether any analytical model or assumptions match the structure.
An unusual reading may reflect an actual structural condition, an environmental change, a sensor problem, a processing error, or a weakness in the model.
The data alone rarely tell the full story.
Digital records may change or become unavailable as systems overwrite data, software is updated, accounts are terminated, equipment is replaced, or outside vendors manage project information.
When litigation is reasonably anticipated, counsel should promptly assess the scope of the client’s preservation duty and work with appropriate technical personnel to preserve relevant electronically stored information. The trigger and reasonable scope of preservation depend on the governing law and the circumstances.
In federal cases, Rule 37(e) of the Federal Rules of Civil Procedure may apply when electronically stored information that should have been preserved in anticipation or conduct of litigation is lost because reasonable steps were not taken and cannot be restored or replaced through additional discovery.
Available measures depend on factors that include prejudice and, for the most serious sanctions, whether the party acted with intent to deprive another party of the information’s use. Rule 37(e) addresses the consequences of losing information, but it does not itself create the duty to preserve.
Depending on the technology and the issues in dispute, relevant evidence may include:
Not every item will be available or within a party’s possession, custody, or control. Some materials may also involve privilege, trade secrets, confidentiality, privacy restrictions, or other discovery limits.
Digital evidence should be collected and preserved in a way that protects its integrity and documents how it was handled. Native files and related system records may help investigators understand how a disputed output was produced, although native production is not required or appropriate in every case.
When discovery involves trade secrets or confidential research, development, or commercial information, a party may seek a protective order under Rule 26(c) of the Federal Rules of Civil Procedure. The party seeking protection generally must establish good cause, and the form of protection will depend on the circumstances.
In federal proceedings, Federal Rule of Evidence 702 requires the party offering expert testimony to show, more likely than not, that the rule’s requirements are met.
The witness must be qualified, and the testimony must help the trier of fact, rest on sufficient facts or data, use reliable principles and methods, and reflect a reliable application of those principles and methods to the facts of the case.
Technology-based opinions may raise several recurring issues under Rule 702, including the following five areas.
An expert may understand the engineering issue but lack the knowledge needed to explain the software, AI model, sensor platform, or drone-processing method.
Depending on the issues, counsel may decide that the case calls for more than one technical discipline. A civil engineer may explain the engineering meaning of an output, while a software, data, or remote-sensing expert explains how the system produced it.
Each expert should remain within a qualified scope, and the combined analysis should connect the technology’s output to the engineering conclusion.
A tool that works in one setting may fail under different site conditions or with lower-quality data.
An expert should consider:
Calling a tool “industry standard” does not prove that it worked reliably in the case. Widespread use or general acceptance may be relevant, but it does not replace an examination of the data, application, and fit.
Even a well-designed system can produce a misleading result when the inputs are incomplete, outdated, biased, or unsuitable for the task.
Experts should identify the data supporting important conclusions, explain the basis for relying on those data, and describe how missing or unavailable information was handled. Rule 703 may permit reasonable reliance on information supplied by others, depending on accepted practices in the expert’s field.
A confident conclusion does not solve a missing-data problem.
An expert’s work may become more vulnerable to challenge when the analysis cannot be tested, evaluated, or reproduced to the extent reasonably possible.
Depending on the method, the expert may need to preserve the software version, data, settings, assumptions, and processing steps. Without enough information about the analysis, the opposing party and the court may have difficulty testing its reliability.
Rule 702 does not impose a universal requirement that every result be reproduced exactly. Some events cannot be recreated, and some accepted analyses involve professional judgment. Still, limited transparency, testability, or reproducibility can raise serious questions about an opinion’s foundation and reliable application.
Even a reliable tool can be used to answer the wrong question.
A model designed for long-term maintenance may not identify the cause of a specific failure. A drone survey taken after repairs may not show an earlier condition, although it may support limited inferences when considered with other evidence. A scheduling tool may identify a pattern without proving causation.
The expert must explain why the method fits the relevant facts, time period, and loss claimed in the case.
These problems do not always lead to complete exclusion. Some flaws may prevent the testimony from satisfying Rule 702, while other weaknesses that do not defeat admissibility may be tested through cross-examination and contrary evidence.
Proposed Federal Rule of Evidence 707 was published in 2025 to address certain machine-generated output offered without an expert witness when comparable testimony from a human witness would be subject to Rule 702.
The public-comment period ran from August 15, 2025, through February 16, 2026. After reviewing the comments, the Advisory Committee on Evidence Rules revised the proposal but decided in May 2026 not to recommend further action at that time. The Committee planned additional study and review before deciding whether the rule should proceed.
The U.S. Courts website continues to list the proposal within the rulemaking cycle originally projected for December 1, 2027. That date is not an expected effective date at this stage because the Advisory Committee has not recommended the proposal for final approval and plans further study.
The proposal would need to complete the remaining federal rulemaking steps, and it could be revised, delayed, or withdrawn.
The proposal’s development highlights a central point: proving that a machine or AI system produced an output does not, by itself, prove that the output is accurate or reliable.
Although the right approach will depend on the technology and the case, experts may consider the following practical investigation framework:
Experts should separate what the technology directly shows from what they infer and what remains uncertain.
The NIST Artificial Intelligence Risk Management Framework may help structure an AI review through its four functions: Govern, Map, Measure, and Manage. NIST describes the framework as voluntary.
It may help organize a technical review, but it does not, by itself, establish the legal standard of care. Its relevance in a particular dispute will depend on the governing law, industry practices, contractual requirements, and expert foundation.
Counsel should determine whether the retained expert or combination of witnesses has the qualifications needed to address both the engineering issue and the technology that produced the disputed evidence.
Counsel may want to ask:
A careful explanation of limitations often strengthens credibility more than an unsupported claim of certainty.
Can an Engineer Avoid Responsibility by Relying on AI?
Using AI does not automatically relieve an engineer of applicable professional duties, even when a software developer or another project participant may also bear responsibility. ASCE Policy Statement 573 states that civil engineers must maintain professional responsibility when AI is used in civil engineering work and that AI cannot replace the professional judgment of a licensed professional engineer. The policy provides professional guidance, but it does not create a uniform rule of law. The engineer’s potential legal responsibility will depend on the applicable licensing and tort law, contract terms, assigned duties, professional standard of care, causation, and the facts surrounding the design.
Are Digital-Twin Records Admissible in Court?
Digital-twin information may be offered in several forms, including underlying records, model files, visualizations, simulations, or expert exhibits. When authentication is required, the proponent must produce evidence sufficient to support a finding that the item is what the proponent claims. Authentication alone does not establish that the information is accurate, reliable, nonhearsay, or admissible for the purpose offered. Other rules may apply depending on the nature of the material and how it is presented, including rules governing relevance, unfair prejudice, hearsay, expert testimony, summaries, illustrative aids, and self-authentication. When an expert relies on a digital twin, the proponent may also need to establish an adequate foundation for the underlying data, model version, methodology, and displayed output. The opposing party may challenge those same issues, including whether the analysis can be tested or reproduced.
When Should Counsel Retain an Engineering Technology Expert?
Counsel may consider early technical assistance when project data may change, repairs could alter physical evidence, or access depends on a proprietary system. Counsel should also decide whether the professional will serve as a consulting expert, testifying expert, evidence-preservation specialist, or another type of technical advisor because those roles may carry different disclosure and work-product consequences. Depending on the case, counsel may seek assistance with technical consulting, evidence collection, investigation, preservation, or expert testimony.
AI can support a range of civil engineering tasks, including data analysis, project planning, monitoring, and risk assessment. Reliability problems can arise when users accept an AI output without checking the underlying data, the system’s intended use and limitations, or the need for professional review.
ASCE Policy Statement 573 states that civil engineers must maintain professional responsibility when AI is used in civil engineering work and that AI cannot replace the professional judgment of a licensed professional engineer.
The policy provides professional guidance rather than a uniform rule of law. Depending on the jurisdiction and the evidence presented, it may still be considered when evaluating professional practices or the applicable standard of care.
An expert who merely repeats an AI-generated conclusion without establishing a reliable foundation may have difficulty satisfying Federal Rule of Evidence 702. That foundation may include the underlying data, validation, system version, intended use, and relevant human oversight.
One potential source of disagreement is version control, especially when project participants rely on different versions of the same model.
Building information modeling is a process for creating, managing, and coordinating digital information about a facility across design, construction, and operations. A digital twin may build on a BIM model by incorporating sensor data, operational information, and current or historical records associated with the physical asset.
Reliability questions may involve outdated versions, missing updates, incomplete sensor data, restricted access to source information, or use beyond the model’s intended purpose.
Detail is not the same as accuracy. An expert should determine whether the model reflects the relevant condition and date.
For example, if an expert cannot identify which model version produced a visualization or explain how missing sensor data were handled, the analysis may be difficult to authenticate, test, reproduce, or defend under Rule 702. Authentication of the item and reliability of the expert’s opinion remain separate legal questions.
Drones are used for surveying, inspections, mapping, documentation, and other technical investigations. A polished image or three-dimensional model may still raise questions about authentication, accuracy, reliability, scale, geolocation, or completeness if the collection or processing steps are poorly documented.
Many nonrecreational civil operations involving small unmanned aircraft systems are conducted under 14 C.F.R. Part 107. Other statutory, regulatory, exemption, waiver, or public-aircraft pathways may apply depending on the aircraft and operation.
Depending on the technical issues in the case, counsel may consider preserving:
Part 107 does not require every item on this litigation-preservation list in every operation. Depending on the dispute, these materials may still help counsel evaluate authentication, reliability, accuracy, and reproducibility.
Compliance with Part 107 also does not, by itself, establish that the resulting imagery or analysis is accurate or admissible.
3D-printed construction raises questions about material consistency, layer bonding, curing, reinforcement, tolerances, testing, and applicable codes or approval requirements.
An investigation may need to examine the printing equipment, material composition or mixture, environmental conditions, digital design files, toolpath instructions, printing sequence, interruption history, reinforcement placement, and available quality-control information.
Because the structure is created layer by layer, problems with material consistency, bonding, or the printing process may weaken structural performance.
Embedded sensors can collect data about structural behavior and surrounding conditions. Conclusions based on sensor data may become less reliable when the data are incomplete, collection conditions are poorly documented, sensors are not properly calibrated, or the analytical model does not reflect the structure.
Depending on the issue, an expert may examine how the data were collected, whether the sensors were calibrated and functioning properly, which conditions were measured, how the data were processed, and whether any analytical model or assumptions match the structure.
An unusual reading may reflect an actual structural condition, an environmental change, a sensor problem, a processing error, or a weakness in the model.
The data alone rarely tell the full story.
Digital records may change or become unavailable as systems overwrite data, software is updated, accounts are terminated, equipment is replaced, or outside vendors manage project information.
When litigation is reasonably anticipated, counsel should promptly assess the scope of the client’s preservation duty and work with appropriate technical personnel to preserve relevant electronically stored information. The trigger and reasonable scope of preservation depend on the governing law and the circumstances.
In federal cases, Rule 37(e) of the Federal Rules of Civil Procedure may apply when electronically stored information that should have been preserved in anticipation or conduct of litigation is lost because reasonable steps were not taken and cannot be restored or replaced through additional discovery.
Available measures depend on factors that include prejudice and, for the most serious sanctions, whether the party acted with intent to deprive another party of the information’s use. Rule 37(e) addresses the consequences of losing information, but it does not itself create the duty to preserve.
Depending on the technology and the issues in dispute, relevant evidence may include:
Not every item will be available or within a party’s possession, custody, or control. Some materials may also involve privilege, trade secrets, confidentiality, privacy restrictions, or other discovery limits.
Digital evidence should be collected and preserved in a way that protects its integrity and documents how it was handled. Native files and related system records may help investigators understand how a disputed output was produced, although native production is not required or appropriate in every case.
When discovery involves trade secrets or confidential research, development, or commercial information, a party may seek a protective order under Rule 26(c) of the Federal Rules of Civil Procedure. The party seeking protection generally must establish good cause, and the form of protection will depend on the circumstances.
In federal proceedings, Federal Rule of Evidence 702 requires the party offering expert testimony to show, more likely than not, that the rule’s requirements are met.
The witness must be qualified, and the testimony must help the trier of fact, rest on sufficient facts or data, use reliable principles and methods, and reflect a reliable application of those principles and methods to the facts of the case.
Technology-based opinions may raise several recurring issues under Rule 702, including the following five areas.
An expert may understand the engineering issue but lack the knowledge needed to explain the software, AI model, sensor platform, or drone-processing method.
Depending on the issues, counsel may decide that the case calls for more than one technical discipline. A civil engineer may explain the engineering meaning of an output, while a software, data, or remote-sensing expert explains how the system produced it.
Each expert should remain within a qualified scope, and the combined analysis should connect the technology’s output to the engineering conclusion.
A tool that works in one setting may fail under different site conditions or with lower-quality data.
An expert should consider:
Calling a tool “industry standard” does not prove that it worked reliably in the case. Widespread use or general acceptance may be relevant, but it does not replace an examination of the data, application, and fit.
Even a well-designed system can produce a misleading result when the inputs are incomplete, outdated, biased, or unsuitable for the task.
Experts should identify the data supporting important conclusions, explain the basis for relying on those data, and describe how missing or unavailable information was handled. Rule 703 may permit reasonable reliance on information supplied by others, depending on accepted practices in the expert’s field.
A confident conclusion does not solve a missing-data problem.
An expert’s work may become more vulnerable to challenge when the analysis cannot be tested, evaluated, or reproduced to the extent reasonably possible.
Depending on the method, the expert may need to preserve the software version, data, settings, assumptions, and processing steps. Without enough information about the analysis, the opposing party and the court may have difficulty testing its reliability.
Rule 702 does not impose a universal requirement that every result be reproduced exactly. Some events cannot be recreated, and some accepted analyses involve professional judgment. Still, limited transparency, testability, or reproducibility can raise serious questions about an opinion’s foundation and reliable application.
Even a reliable tool can be used to answer the wrong question.
A model designed for long-term maintenance may not identify the cause of a specific failure. A drone survey taken after repairs may not show an earlier condition, although it may support limited inferences when considered with other evidence. A scheduling tool may identify a pattern without proving causation.
The expert must explain why the method fits the relevant facts, time period, and loss claimed in the case.
These problems do not always lead to complete exclusion. Some flaws may prevent the testimony from satisfying Rule 702, while other weaknesses that do not defeat admissibility may be tested through cross-examination and contrary evidence.
Proposed Federal Rule of Evidence 707 was published in 2025 to address certain machine-generated output offered without an expert witness when comparable testimony from a human witness would be subject to Rule 702.
The public-comment period ran from August 15, 2025, through February 16, 2026. After reviewing the comments, the Advisory Committee on Evidence Rules revised the proposal but decided in May 2026 not to recommend further action at that time. The Committee planned additional study and review before deciding whether the rule should proceed.
The U.S. Courts website continues to list the proposal within the rulemaking cycle originally projected for December 1, 2027. That date is not an expected effective date at this stage because the Advisory Committee has not recommended the proposal for final approval and plans further study.
The proposal would need to complete the remaining federal rulemaking steps, and it could be revised, delayed, or withdrawn.
The proposal’s development highlights a central point: proving that a machine or AI system produced an output does not, by itself, prove that the output is accurate or reliable.
Although the right approach will depend on the technology and the case, experts may consider the following practical investigation framework:
Experts should separate what the technology directly shows from what they infer and what remains uncertain.
The NIST Artificial Intelligence Risk Management Framework may help structure an AI review through its four functions: Govern, Map, Measure, and Manage. NIST describes the framework as voluntary.
It may help organize a technical review, but it does not, by itself, establish the legal standard of care. Its relevance in a particular dispute will depend on the governing law, industry practices, contractual requirements, and expert foundation.
Counsel should determine whether the retained expert or combination of witnesses has the qualifications needed to address both the engineering issue and the technology that produced the disputed evidence.
Counsel may want to ask:
A careful explanation of limitations often strengthens credibility more than an unsupported claim of certainty.
Can an Engineer Avoid Responsibility by Relying on AI?
Using AI does not automatically relieve an engineer of applicable professional duties, even when a software developer or another project participant may also bear responsibility. ASCE Policy Statement 573 states that civil engineers must maintain professional responsibility when AI is used in civil engineering work and that AI cannot replace the professional judgment of a licensed professional engineer. The policy provides professional guidance, but it does not create a uniform rule of law. The engineer’s potential legal responsibility will depend on the applicable licensing and tort law, contract terms, assigned duties, professional standard of care, causation, and the facts surrounding the design.
Are Digital-Twin Records Admissible in Court?
Digital-twin information may be offered in several forms, including underlying records, model files, visualizations, simulations, or expert exhibits. When authentication is required, the proponent must produce evidence sufficient to support a finding that the item is what the proponent claims. Authentication alone does not establish that the information is accurate, reliable, nonhearsay, or admissible for the purpose offered. Other rules may apply depending on the nature of the material and how it is presented, including rules governing relevance, unfair prejudice, hearsay, expert testimony, summaries, illustrative aids, and self-authentication. When an expert relies on a digital twin, the proponent may also need to establish an adequate foundation for the underlying data, model version, methodology, and displayed output. The opposing party may challenge those same issues, including whether the analysis can be tested or reproduced.
When Should Counsel Retain an Engineering Technology Expert?
Counsel may consider early technical assistance when project data may change, repairs could alter physical evidence, or access depends on a proprietary system. Counsel should also decide whether the professional will serve as a consulting expert, testifying expert, evidence-preservation specialist, or another type of technical advisor because those roles may carry different disclosure and work-product consequences. Depending on the case, counsel may seek assistance with technical consulting, evidence collection, investigation, preservation, or expert testimony.
What Methodology Problem Commonly Arises in Technology-Based Testimony?
A common problem is relying on software output without establishing a reliable basis for the inputs, assumptions, validation, limitations, and connection to the claimed loss. When information is missing or unavailable, the expert should identify the gap, explain the basis for relying on the information that is available, and describe how the limitation affects the opinion.
Construction Engineering
Digital Modeling
Unmanned aerial vehicle uav
Aerial Mapping
Light detection and ranging lidar
3D Printed Homes
Sensors
Sensor Processing
Digital Evidence
Data Analysis
Expert Witness Insights by Area of Expertise
Expert Witness Insights by Area of Expertise
What Methodology Problem Commonly Arises in Technology-Based Testimony?
A common problem is relying on software output without establishing a reliable basis for the inputs, assumptions, validation, limitations, and connection to the claimed loss. When information is missing or unavailable, the expert should identify the gap, explain the basis for relying on the information that is available, and describe how the limitation affects the opinion.