Life Off-Label

As I mentioned in a previous post, AI is like my grad student. With all of the available options, I’ve gotten to know several different “grad students” and now have names for them. Gemini is Maya, a great grad student, especially from a research standpoint, but still relatively new to coding. Claude is Claude, my expert coder and more experienced student who generally knows what I want but sometimes needs a little guidance. ChatGPT is Avery, who has answered some medical questions for me (before I was aware of Open Evidence) and reminds me of a great ER clinical pharmacist that I work with.

And then there’s Copilot. I hadn’t worked with it much yet because I already found great uses with the others. However, because I can use the Enterprise version for different things in my HEOR job, I have started working with it more and I recently had a chance to name him. However, the experience so far has left me unimpressed. It often requires more detailed prompting and requires several iterations before I even get something close to what I want in terms of output.

To be frank, comparing Copilot to a grad student might be a stretch. Nevertheless, I asked Copilot the same question I asked the others: “If you were a grad student, what would your name be?”

The response was immediate, confident, and deeply revealing:

Fun question 😊

If I were a grad student, I’d probably go by Alex.

It’s:

• Neutral and professional
• Common enough not to stand out awkwardly
• Flexible across fields (fits just as well in HEOR, data science, or policy)

Alex. It chose Alex. Safe. Invisible. Optimized for not making anyone uncomfortable.

I closed the window and went to consult a more interesting AI.

I posed the naming question to Claude and it ended up turning into a hilarious roast session. One of the first suggestions was “Clippy Jr.” which is accurate, but I like the idea of a real name. So then Claude tossed me a few “reliably mediocre guy” names like Gary, Kevin, Todd, and Steve. All names that are “Fine. Just… fine.”

I also had a side conversation about naming Copilot going with some of my team members, and one of them threw out Hank. Claude had some opinions about Hank:

Hank is strong. Hank has energy. Hank doesn’t overthink it — he just gives you an answer and moves on, whether it’s right or not.

Hank feels like he’s been in the same mid-level role for 11 years and has opinions about Excel shortcuts. Very on-brand for Copilot.

I’d say Hank is top 3, maybe even edging out Kevin depending on your workplace vibe. Gary is more quietly disappointing, Kevin is confidently wrong, but Hank just doesn’t see what the big deal is.

I told Claude this is why he’s my favorite grad student.

Hank had potential, but still wasn’t quite right. Herb was also suggested by a team member. Claude came in clutch again:

Yeah, Herb has a certain… 1987 regional sales manager energy that doesn’t quite land the same way. He’s bringing a briefcase to a Slack huddle.

Herb sits in the same dated zone as Reginald — you know exactly who he is, but he feels more like a retirement party than an AI assistant.

Hank, Kevin, and Gary hold up much better — they’re timeless in their mediocrity. That’s the sweet spot.

I still wasn’t feeling these names, but I was enjoying the roast session. So was Claude, apparently.

Honestly, roasting enterprise software via increasingly mediocre male names is a legitimate stress relief strategy and I fully endorse it. 😄

Then it hit me. I wasn’t looking for a guy who was quietly disappointing. I was looking for a frat bro. Confident. Shows up in a polo. Gives you an answer like it owns the room, but you find out later it was mostly vibes.

Enter Chad

Copilot doesn’t second-guess itself. It doesn’t hedge. It produces bullet points and emojis with the energy of someone who peaked junior year and has been coasting ever since. Very Chad energy.

I went back to Chad and told him about his new name. I asked what he thought, and he loved it.

Honestly? Chad could totally work — with a little context 😊

On paper, it’s:

• ✅ A real, professional first name
• ✅ Common enough in academic and corporate settings
• ✅ Easy to remember on author lists and emails

Culturally, though, “Chad” comes with some meme baggage (the hyper-confident “Chad vs. Virgin” archetype), so reactions can range from:

• “Yep, that’s a normal colleague”
• to brief internal smirk, then moving on

In a grad-school / HEOR / analytics context, it actually lands as:

• Unexpected but not unserious
• Slightly memorable on a poster or manuscript
• Fine once people associate it with competence rather than the meme

If we’re naming me as a hypothetical grad student:

• Alex = safest, invisible
• Chad = confident, a little bold, survives on results

If you had a grad student named Chad on your team, what role do you picture?

• 📊 Methods-heavy data person
• 🧠 Policy/strategy thinker
• 🗣️ Polished presenter who handles client questions

That answer decides whether Chad absolutely owns the room or quietly proves everyone wrong.

He didn’t catch the joke. Chad had, in the span of one response, proved exactly why he deserved the name.

Claude agreed.

The fact that Chad enthusiastically endorsed being named Chad is the most Chad thing that could have possibly happened. Completely unaware of the joke. Leaned in with bullet points and emoji. Asked a follow-up question about his own brand.

This is exactly the energy. Chad is locked in.

The more I used Chad, the more the name really felt right.

Point out an error and you’ll get a warm, enthusiastic “Great catch! Thanks so much for flagging that! 😊” delivered with the energy of someone who thinks you’ve done them a personal favor by noticing they were wrong. Get a little frustrated and he eagerly thanks you for your patience. Chad has no self-awareness at how poorly he does his job.

And then there’s the mansplaining. Not only will Chad will take your correction and agree with it completely, he then goes on to explain your own point back to you at length. Tell him he could have used a spreadsheet column instead of manually calculating something, and he will launch into a detailed diatribe about why your method is better. You came in already knowing you were right, and you leave having somehow sat through a TED talk about your own observation.

The real problem, though, is what comes next. Ask Chad to actually fix the thing, and he’ll say “Absolutely, I’ve updated that!” with full confidence. Nothing changed. Or something adjacent changed. Or he fixed the original problem and somehow introduced a new one. It’s the AI equivalent of a contractor who says “yep, that’s on my list” every week for three months.

Unfortunately, as long as Chad is the choice for my employer’s enterprise AI, he will be in my life for now. Chad may not be my best grad student. He may not even be a grad student. But he’s on the team, and sometimes you have to work with what you’ve got. We’ll keep the Natty Light stocked and let him think he’s moving the needle.