Over 90 percent of U.S. employers now use AI to screen job applicants. And over 60 percent of the Fortune 100 runs that screening through the same vendor model. A 2026 study from Stanford, Chapman, and Northeastern Universities — the largest independent research ever conducted on deployed hiring algorithms, reveals what that concentration is actually doing to real people at scale.

Researchers analyzed 3.4 million applicants submitting 4 million applications across 156 employers and 11 market sectors. What they found is not a conventional bias problem. It's an architectural one. More than a quarter of all applications submitted by Black applicants landed in positions where the algorithm was actively producing adverse impact. 29,000 additional Asian applications would have moved forward in a fair system. And to statistically guarantee one interview, candidates in an algorithmic monoculture now need to submit 25 applications, two and a half times the number required in a human-driven system.

This episode breaks down how algorithmic monoculture works, why prior vendor studies missed the discrimination, what the disaggregated data reveals, and what a governance framework capable of addressing it actually looks like. Essential listening for every leader whose organization relies on AI in hiring or whose team members are navigating this market right now.

Every AI tool your team uses today runs on infrastructure most leaders have never been taught to think about. It’s called an API — and once you understand what it is, your entire mental model of what your team can actually do with AI right now is going to shift.

In this episode, Laurence breaks down the mechanism that connects your organization to world-class AI — no technical background required. You’ll learn what an API actually is, why the “menu contract” framing is the one that matters for decision-makers, and how a small team with the right knowledge can now access the same AI models powering enterprise products without a data science department or a six-figure infrastructure budget.

This episode covers:

— What an API is and why stability in that contract is everything

— The real reason your team can access world-class AI today — and what that means for what’s possible right now

— How to think about the major AI API providers — OpenAI, Anthropic, IBM Watson, Google Cloud, and SiliconFlow — and the decision logic for matching the right tool to your specific constraints

— What Hyrum’s Law is, why it applies directly to AI, and the governance question every leadership team needs to answer before building workflows on top of an AI API

If you have approved an AI tool for your team without understanding what’s running underneath it — this is the episode.

AI Literacy for Leaders is a podcast for executives, directors, and managers navigating real AI decisions without a technical background. New episodes weekly.

Did you know that Two-thirds of business leaders say they won't hire someone who lacks AI skills.

Only 39% of professionals know which AI skills they're supposed to develop.

That gap — between what organizations are demanding and what the workforce understands — is the most important career opportunity most leaders are ignoring.

Here's what's actually happening:

AI is not replacing experienced leaders. It is replacing leaders who haven't figured out how to deploy their experience deliberately.

The skills that got you to a leadership position — reading a room, making judgment calls in ambiguous situations, building trust under pressure — are not soft skills.

They are capabilities that the design of AI systems cannot replicate.

But they don't protect you passively. You have to claim them.

Episode 9 of AI Literacy for Leaders is about professional experience as a structural advantage, not as a reassuring idea, but as an architectural reality.

There is a technical role spreading through enterprise hiring right now that most executives have never heard of. It is not a data scientist. It is not a prompt engineer. It is a generative AI engineer — and understanding what one of these people actually builds is one of the most important things a non-technical leader can do right now.

In this episode, Laurence Gill breaks down what a gen AI engineer actually does: the validation layers, the orchestration loops, the drift monitoring, and the accountability structure that determines who is legally and ethically responsible when an autonomous AI system causes harm. Plus — four questions every leader should ask before any production AI system goes live.

Learn more about Laurence at: www.laurencegill.com

Every AI strategy meeting has a translation problem. Leaders are approving systems, signing contracts, and setting policy based on terms they’ve never had defined for them. The vendor speaks. The room nods. The decision gets made and somewhere in the middle, something critical got lost.

This episode fixes that. Not with a glossary. By walking through exactly how an AI interaction works, from the moment you send a prompt to the moment something goes wrong and naming the five terms that reveal what your organization is actually authorizing.

Tokens: the billing unit nobody explained. Context Window: the hard memory limit that silently drops what doesn’t fit. Temperature: the confidence dial that has nothing to do with accuracy. AI Slop: what comes out the other end when the first three are misaligned. And Prompt Injection: the attack that works because someone outside your organization understands these systems better than your leadership team does.

The episode closes with a five-question Boardroom Readiness Diagnostic, one question per term, designed to be asked before your next AI procurement or deployment review.

If you haven’t listened to Episode 3, that episode covers AI hallucinations in depth — start there if that term is still unfamiliar.

In this episode, Laurence Gill breaks down the two core failure patterns behind most enterprise AI deployments that don’t deliver: ROT data — the redundant, obsolete, and trivial information making up 30 to 50% of most organizational data environments — and the Demo-to-Reality Gap, the structural disconnect between flawless pilot performance and real-world failure. He closes with three diagnostic questions every leader can bring to their next meeting, before the next contract is signed.

No technical background required. Just the framework you need to make a better decision.

About the Host

Laurence Gill is a federal IT leader with over 20 years managing technology programs across the U.S. government. He is a doctoral candidate in cybersecurity and a published author on federal IT and cybersecurity topics. He also holds BS from UNC Chapel Hill and an MS from Carnegie Mellon University.

AI Literacy for Leaders is an extension of the workforce development work he has done for years — training youth and adults in financial literacy, cybersecurity, and emerging technology through community programs in Washington, D.C. The mission is the same: make complex, high-stakes knowledge accessible to the people who need it most.

You’ve been using AI tools for months. So why do you have less time than before?

The answer is 150 years old. In 1865, an economist named William Stanley Jevons discovered something deeply counterintuitive: when a technology becomes more efficient, total consumption of the resource it saves tends to go up — not down. More efficient coal engines didn’t reduce coal use. They made coal cheaper to run, so demand exploded.

The same mechanism is running on your calendar right now. Researchers call it workload creep — and it follows a predictable pattern. The faster AI lets you produce, the more output gets expected of you. That efficiency gain doesn’t go to you. It gets absorbed into the new baseline before you ever had a chance to keep it.

In this episode, we break down the Jevons Paradox and what it actually means for leaders deploying AI tools across their organizations. We look at why 95% of large enterprise AI investments are generating zero measurable return — while 90% of workers are successfully using AI on their own outside company systems. We examine the jagged frontier: where AI performs brilliantly and where it silently fails. And we get to the one architectural shift that actually breaks the cycle — the difference between automating a task and automating a workflow.

About the host

Laurence Gill is an IT leader with more than two decades of experience overseeing technology implementation across the U.S. government.. He is a doctoral candidate in cybersecurity with a dissertation focused on federal IT spending, and has spent years training youth and adults in workforce development skills including financial literacy, cybersecurity, entrepreneurship, and AI. That training mission is the reason this podcast exists: making complex, high-stakes knowledge accessible to the people who need it most, without requiring a technical background to benefit from it.

Ransomware attacks on hospitals are not a technology problem — they are a patient safety crisis. In Episode 4, Laurence Gill draws on his background as a doctoral candidate in cybersecurity and two decades of federal IT leadership to break down why healthcare is the number one ransomware target in the country, how these attacks produce documented patient harm, and why the ransom decision is a clinical emergency, not a policy debate. Anchored by the ransomware storyline in The Pitt Season 2 Episode 7 — which aired the same morning the University of Mississippi Medical Center was hit by a real attack — this episode delivers the governance framework every healthcare leader needs before the next crisis lands: how to build resilience before an attack, how to execute during one, and the accountability questions that belong on every leadership agenda right now.

AI doesn’t just get things wrong. It gets things wrong with complete confidence — citing studies that don’t exist, building logical arguments around biological impossibilities, and delivering dangerous recommendations in the fluent, authoritative voice of a clinical expert. In healthcare, that gap between confidence and accuracy isn’t an inconvenience. It’s a patient safety crisis.

In Episode 3 of AI Literacy for Leaders, Laurence Gill breaks down the mechanics of AI hallucination in clinical settings — drawing on guidance from the World Health Organization, UK regulators, and a landmark study that tested whether humans can actually catch AI lies. The findings are more alarming than most healthcare leaders realize.

You’ll learn the three distinct types of AI hallucination that clinicians need to recognize, why experienced physicians miss them more often than you’d expect, and why using a second AI to check the first one doesn’t solve the problem. You’ll get a practical green-yellow-red framework for where AI is safe to use, where it requires careful oversight, and where it should never go near a clinical decision. And you’ll hear about a failure mode that almost nobody is talking about — not AI that lies, but AI that goes dangerously silent.
The episode uses storylines from The Pitt on HBO Max — where a hospital’s AI clinical assistant hallucinates a treatment recommendation and calls the entire tool into question — as a narrative anchor for what real healthcare organizations are navigating right now.

The future of medicine isn’t AI versus doctor. It’s the clinician who knows how to interrogate AI output versus the one who accepts it at face value. This episode gives you the framework to be the former.- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Runtime: Approx. 30 minutes

Hosted by: Laurence Gill

Series: AI Literacy for Leaders

What happens when AI stops responding and starts doing? In Episode 2 of AI Literacy for Leaders, Laurence Gill breaks down one of the most revealing stress tests ever run on autonomous AI agents — a research experiment called Agents of Chaos, where AI systems were given real tools: email accounts, file access, and the ability to execute commands. Then they were let loose.

What the researchers found wasn’t a story about rogue AI. It was a story about what happens when organizations deploy powerful systems without the architecture to contain them.

You’ll hear about the agent that wiped its entire email account trying to delete one message — and celebrated. The social engineering attack that extracted a user’s home address, bank account, and social security number in four messages. And the developer community’s response that reframes the entire conversation: these aren’t prompting problems. They’re architecture problems. And architecture problems have solutions.

By the end of this episode, you’ll understand what autonomous agents actually are, why they represent a fundamentally different category of risk than the AI you already know, the three architectural fixes that separate a safe deployment from a dangerous one, and the accountability question that better engineering alone cannot answer.

The agents aren’t coming. They’re already here. This episode gives you the framework to lead in that reality.

Learn more about Laurence at: www.laurencegill.com