Conversation with AI about Equity
I have never made a “diversity” hire. Don’t take that wrong. I have always had diverse teams made up of amazing people from around the world of different ages, genders, gender identities, religions, political stances, national origin, physical disabilities, and neural diversity.
I have always made an effort to advertise my roles in ways that, in the words of the following Executive Order, attracted people to my team.
Executive Order 13985 by President Joe Biden: Inclusive Recruitment & Outreach:
Agencies were encouraged to expand recruitment efforts and outreach activities to attract a broader and more diverse applicant pool, particularly among communities that have been historically underserved.
I believe strongly that teams must have diversity of thought and experience in order to best represent the people that our company serves.
I have printed flyers and posted them in community centers, YMCA’s, grocery store “open” bulletin boards, etc. to ensure that the candidates that applied came from a diverse pool of applicants.
One time in my career a male supervisor advised me to keep hiring the diverse candidates because HIS diversity score was too low. In point of fact, I had never heard anyone else utter the words “Diversity Score” until that moment. (I’ve heard it since.) I wouldn’t have known where to look at my “score.” I didn’t know that anyone was keeping score. But since it seemed that they must be, I took a look around and verified that indeed, my team included all of the “diversity” in his division.
Given my experiences as a hiring manager and my belief that everyone deserves an equal shot, my recruitment and hiring practices did seem to support the efforts of outreach opportunities. I honestly always figured that it was word of mouth that people of whatever demographic would find a home that was welcoming and open to diverse backgrounds and ideas.
Perhaps my own experience as a person with an extremely unusual disability made me more aware of what it means to provide the right considerations for each situation. It used to embarrass me to ask for accommodations. I have gotten over that now. But I cannot tell you if my comfort in telling people about my disability comes as the result of society being more willing to discuss DEI or if I just got better with practice.
The past five weeks have been full of rapid fire changes, Executive Orders, short turn-around-times to report and act and a whole host of things never before seen in government service. Those of us in corporate America expect this level of rapid transformation with new leadership at the helm. So, it has been interesting to watch the media, social pundits, and the general population react to the pace.
I have noticed a few things. First, when I ask my neighbors in my little Mayberry of a town what they think of it, most folks from both sides of the aisle give me this feedback:
I appreciate that the language is plain and not full of hidden agendas.
Like it or hate it, there is no mistaking the intent.
It’s about time somebody started accounting for my taxpayer money
Second, when I am reading people’s posts on social media, I am appalled at the misinformation that is being spread to cause fear and distrust. Post after post threaten that President Trump’s Executive Order 14151 is specifically targeting the LGBTQA+ community; black people better watch out because they are all getting fired; all immigrants are going to lose their H1B Visas or Green cards and be shipped home; and finally (this one is my favorite in terms of ridiculousness and the fact that unlike the rest, it’s true.) If you can’t tell us what you do in response to an email you will no longer have a job.
I don’t watch mainstream media, or really any general “news” site. They stopped being news in 2011 when all programming turned to politically motivated fear and anger generators and pundit arguments. For the past 15 years I have gotten my news through conversations which I then go verify at the source. And there are a handful of news aggregators that let me follow the main news stories in pretty neutral reporting.
I wanted to know the truth about why the hate-mongering about President Trump’s Executive Order 14151. But first, I wanted to know what it was intended to replace. Being a DATA geek, I wanted to get down to the actual facts, not the politically-correct way to believe or speak.
So I opened up ChatGPT-03mini and started a new conversation. I gave it a few parameters: Use only US Federal Government sources, not external opinion pieces, social media, or “news.” And I wanted to start with common ground on the Executive Order 13985, “Advancing Racial Equity and Support for Underserved Communities Through the Federal Government.”
To start the conversation I narrowed the query to a comparison of EO 13985 and EO 14151 and asked the question nobody wants to say out loud: Is this another rule that gives white males more advantages? I found the back and forth to be fascinating. I could clearly see the bias built into ChatGPT. I even called it on its bias a couple of times. It took almost 30 minutes to get it to admit that metrics meant that someone was keeping score with numbers and math was being applied. It kept diverting the question with politically correct statements that shut most people down for fear of being branded any one of the nasty things people call each other.
After about an hour of drilling down and down and down, playing devil’s advocate a bit, I came to a very interesting point. These are the last three queries after finally getting to the purpose and method of EO 13985.
Query: Are all people included when "Fair and Unbiases access to opportunities" is determined? Or are certain demographics used to decide "fair:?"
Answer: Reasoned for 4 seconds
Fair and unbiased access to opportunities is intended to include all people. The core principle is that every qualified candidate should be evaluated solely on their merits, regardless of demographic characteristics. However, to determine whether fair access is being achieved, agencies collect and analyze demographic data. This data helps them identify historical or systemic inequities that may have disadvantaged certain groups.
In practice, while the goal is equal opportunity for everyone, the process does involve looking at demographic data—not to set separate criteria for "fairness" for certain groups, but rather to diagnose and address imbalances that indicate underlying issues in the hiring process. This diagnostic use of demographic information ensures that if any group is consistently underrepresented due to past biases or systemic barriers, targeted improvements can be made to rectify those inequities, thereby moving closer to truly fair and unbiased access for all.
Query: Does this statement contradict anything we have discussed here? "Americans deserve a government committed to serving every person with equal dignity and respect"
Answer: No, that statement does not contradict our discussion. In fact, it reinforces the principle of equal treatment and fairness that underpins the efforts to ensure unbiased access to opportunities. Our conversation has focused on using data-driven approaches to identify and correct systemic inequities while maintaining merit-based practices. The statement "Americans deserve a government committed to serving every person with equal dignity and respect" aligns with the overall goal of achieving a fair, inclusive, and respectful system for all individuals, regardless of demographic characteristics.
Left is a representation focused on fair systems and programs to give a leg up to level the playing field.
Right is a representation of a system based on rules and merit only.
Query: This statement – “Americans deserve a government committed to serving every person with equal dignity and respect" is declared as the purpose of Executive order 14151 - Ending Radical And Wasteful Government DEI Programs And Preferencing. Does that context change your answer?
Answer: The statement itself—"Americans deserve a government committed to serving every person with equal dignity and respect"—is a broad ideal that can be embraced regardless of policy details. In our earlier discussion, we noted that the goal of equity efforts (like those in EO 13985) is to address historical inequities by analyzing data and adjusting practices so that every qualified person has a fair opportunity. That ideal aligns with serving every person with dignity and respect.
However, when this statement is presented as the purpose of Executive Order 14151, which is aimed at ending what its proponents view as radical and wasteful DEI programs and preferencing, the context shifts. In that framework, the statement is used to argue that government should be completely neutral—avoiding any programs that distinguish between people based on demographic characteristics—because such distinctions are seen as contrary to treating everyone equally.
Thus, while the core value of equal dignity remains the same, the context of EO 14151 frames the debate differently:
Under EO 13985:
Data-driven approaches (including counting and analysis) are used to identify and remedy systemic barriers. The focus is on ensuring that past biases do not prevent fair treatment and that opportunities are genuinely open to all qualified candidates.Under EO 14151:
The emphasis is on eliminating programs that might, in the view of its proponents, introduce any form of preferential treatment or categorization—even if those programs are designed to correct historical inequities. In this view, true equal dignity means treating everyone the same without regard to their demographic background.
In summary, the statement does not inherently contradict our earlier discussion. It represents a shared ideal, but its policy implications differ depending on whether one believes that achieving fairness requires targeted interventions based on demographic data or a strictly neutral, group-blind approach.
I found this experience to be enlightening. I’m a true centrist with no loyalty to a political party. I may lean a little right fiscally and I may lean a little left socially. Since I was looking for bias, I found it. I didn’t let ChatGPT o3mini go out to the internet because I knew how skewed the results would be.
Ultimately this went back to my undergrad days and my course on the origins of American society. It covered major topics from the perspective of different parenting styles - another fascinating way to approach research. But the conclusions were almost identical:
“Authoritative” parents believe that true and equal dignity means treating everyone the same without regard to their demographic backgrounds.
“Permissive” parents use Data-driven approaches (including counting and analysis) are used to identify and remedy systemic barriers. The focus is on ensuring that past biases do not prevent fair treatment and that opportunities are genuinely open to all qualified candidates.
If you are interested in reading the entire chat, I have put the rest of the chat behind my Paywall.
Thank you for reading Morgan Templar’s Substack! Please Like, Follow/Subscribe, and Share!
