Eric Vaughan

One question. Disqualifies more candidates than anything else I ask.

"How did you prepare for today?"

Website scrolling and Google searches is the wrong answer. Not because it's incompetent. Because someone who instinctively uses AI would have fed this company into Perplexity before ever walking in the door. Products, press releases, sectors, background on who they're meeting. For a job they want.

The instinct gives you away before the interview really starts.

From my conversation with Taylor Malmsheimer on Section's Engine of AI.

A year in, I figured out what actually separated the people who made it from the people who didn't.

Not skill. Not effort. Belief.

And looking back, I would have found a faster way to divide believers from non-believers from the start. Because spending twelve months with someone who's trying without believing is a waste of everyone's time, including theirs.

I don't have a clean prescription for how to find it earlier. But I know now that it was the only variable that actually mattered.

From my conversation with Taylor Malmsheimer on Section's Engine of AI.

A lot of candidates know AI matters. Fewer have actually built the habit.

That gap shows up fast. The ones who've done the work can tell you exactly what's on their tool belt, what they reach for and why, with the fluency of someone who does it every day. The ones who haven't give you intention. Good answers about what they plan to do. Nothing from what they've actually done.

Aspiration isn't DNA. Evidence is.

From my conversation with Taylor Malmsheimer on Section's Engine of AI.

The company furthest ahead is the company most at risk of stopping.

Two years into this transformation, ahead of most organizations by a wide margin, and my biggest concern is complacency. Not technology. Not cost. Complacency. Because being better than average is not the same as being at the frontier. We should be the leading voice by now. When I catch us following instead of leading, that's embarrassing.

The second thing we battle every day: the AI is built to agree with you. Not to be honest with you. To satisfy you. Use multiple LLMs, push back, hold it accountable, set standards explicitly. The era of obvious hallucinations is mostly behind us. The era of subtle bias shaped by your own framing is very much here.

Both problems are leadership problems. Neither solves itself.

From my conversation with Taylor Malmsheimer on Section's Engine of AI.

If you think you're behind, good. If you don't think you're behind, you're doomed. That's the line I keep repeating, because it's true.

The companies I worry about are treating AI like a side project: minor investment, no CEO buy-in, and the comfortable belief they've got time. Large public companies that feel impervious are making the same mistake.

It's going to come down to AI DNA. The companies that build it in will consolidate the market. The ones that don't will be consolidated.

From my conversation with Nathan Labenz and Prakash Narayanan on AI in the AM.

We didn't post job listings. We sent a signal.

Every open role at IgniteTech carried the same prefix: AI Innovation Specialist. Finance. Marketing. Engineering. Every function. Because the title told candidates exactly who we were looking for before they read a single bullet point.

It worked. We pulled people out of companies where they were being told not to use AI. People with belief and skills and nowhere to put either. We gave them a place.

Then I interviewed every single one of them myself. After rebuilding the company, I wasn't leaving the bar up to chance.

You can develop skill. You cannot install belief. We hired for the one you can't teach.
From my conversation with Taylor Malmsheimer on Section's Engine of AI.

My advice to enterprise on AI is aggressive. Prakash asked me why. My answer: fear.

Decades as a technologist taught me the fear is specific. Someone finding me in a year or two, asking why I didn't warn them their profession was about to be rewritten. I don't want that conversation.

Tech has cried wolf before: crypto, blockchain, Web3. None of it changed everything. This one's not like that.

From my conversation with Nathan Labenz and Prakash Narayanan on AI in the AM.

I couldn't tell if I was measuring aptitude or attitude.

After a full year of investment, unlimited tools, and complete clarity about the direction, some employees still weren't engaging. Hard to separate "can't" from "won't" when someone isn't really trying. Turned out it didn't matter much. Neither was good. Attitude is just worse. Because attitude is a choice.

We put it on a bar chart. Three zones. Talked to everyone in the basement one on one. Not to shame anyone. To be honest: we want you to stay, but you've got to do your part.

Most leaders make that conversation vaguer than it has to be. We didn't.

From my conversation with Taylor Malmsheimer on Section's Engine of AI.

We didn't wake up one day and exit 80% of our people. That's not the story, and the real one matters more.

A full year of training, funding, tools, and incentives came first. What followed wasn't a layoff. It was a replacement, people who refused the work swapped for people who wanted it.

Picture a World Cup coach calling a formation and two players saying they don't believe in it. You wouldn't put them on the pitch. We measured effort, not aptitude.

From my conversation with Nathan Labenz and Prakash Narayanan on AI in the AM.

Two things separate the companies winning in AI from the ones writing task forces about it.

The CEO is visibly worried. Not panicked. Worried. Because if you think you're behind, good. That's the correct response to a shift this size. If you don't think you're behind, you're really doomed.

And the tools are unrestricted. No policies drafted by people who don't understand the technology. No locking it down because it's unfamiliar. Your people can't come with their best thinking if you've decided the risk of trying outweighs the risk of standing still.

Strategy, not a tactic. The direction, not a project.

From my conversation with Taylor Malmsheimer on Section's Engine of AI.