Step 1: Contextual Framework Pass (Facts)

Section Content:

Facts:
Engineer A, an environmental engineer with several years of experience and holding a professional engineering license, was retained by Client W to prepare a comprehensive report addressing the manufacture, use, and characteristics of an organic compound identified as an emerging contaminant of concern.
This work required Engineer A to perform an analysis of groundwater monitoring data from a site Engineer A had been observing for over a year.
In addition, Engineer A was tasked to develop engineering design documents (plans and specifications) for modifications to groundwater infrastructure at the same site.
Engineer A is known for their strong technical expertise but is, personally, less confident in their technical writing.
Previously, Engineer A had relied on guidance and quality assurance reviews by their mentor and supervisor, Engineer B, to refine report drafts.
But Engineer B recently retired and was no longer available to Engineer A in a work capacity.
Faced with the need to deliver both the report and the engineering design documents without the review by and mentorship from Engineer B, Engineer A opted to use open-sourced artificial intelligence (AI) software to create an initial draft of the necessary report and to use AI-assisted drafting tools to generate preliminary design documents.
The AI drafting software was new to the market and Engineer A had no previous experience with the tool.
The AI drafting software was also open-sourced.
For the report, Engineer A gathered the relevant information provided by Client W and relied on the AI software to synthesize the information and generate an initial draft of the necessary report.
Engineer A input the information gathered from Client W into the AI software, and, after a few refining prompts, received a first draft of the report generated by the AI software.
Not being familiar with the full functionality of the AI software, including the accuracy and originality of AI-generated text, Engineer A conducted a thorough review of the report, cross-checking key facts against professional journal articles and verifying the phrasing by running search engine queries to ensure the content did not match any existing language.
Engineer A also made minor adjustments to some of the wording to personalize the content.
Engineer A did not cite their use of AI-software or its large language models, and submitted the draft report to Client W for review, including language to clearly identify that the supplied report was a draft, but applied their seal consistent with state law.
For the engineering design documents, Engineer A entered the information gathered from Client W into the AI software and relied on the AI-assisted drafting tools to generate a preliminary design of the plans, including basic layouts and technical specifications.
Engineer A completed a cursory review of the AI-generated plans and adjusted certain elements to align with site-specific conditions.
Again, Engineer A did not cite the AI-assisted drafting tools they used to generate the engineering design documents.
When Client W reviewed the draft report, Client W noted that the section analyzing the groundwater monitoring data would benefit from minor edits for grammar and clarity, but found the introduction discussing the contaminant’s manufacture, use, and characteristics to be exceptionally polished.
The Client commented that the report read as if written by two different authors but was otherwise satisfactory.
Client W, however, noticed several issues with the AI-generated design documents, including misaligned dimensions and an omission of key safety features required by local regulations.
Client W raised concerns about the accuracy and reliability of the engineering design and instructed Engineer A to revise the plans, ensuring that all elements satisfied the necessary professional and regulatory standards.