AI Insights with Micah Gaudet

Latest Episodes:

The Barriers to AI Adoption Jul 04, 2026

Show notes

A LinkedIn thread asking one simple question turned into one of the most honest conversations on AI adoption in local government. Local government professionals, GovTech founders, and vendors weighed in, and nearly every barrier they named traced back to the same root: a lack of clarity, not a lack of technology.

This episode breaks down what that thread revealed: misunderstanding driving fear, employees quietly using AI without admitting it, leaders who limit AI through policy rather than confront their own knowledge gaps, and the difference between responsible governance and avoidance dressed up as caution. It also covers a problem hiding underneath the whole conversation, and a more hopeful case for what AI actually frees local government workers to do.

Source article: I Asked Local Government Professionals About the Biggest Barrier to AI AdoptionFind my work at micahgaudet.comConnect with me on LinkedIn

We have an AI Policy Problem Jul 04, 2026

Show notes

Most AI policies look like governance but don't function as one. This episode traces the problem back to its roots: AI policy inherits its DNA from IT acceptable-use policies and cybersecurity frameworks, built for tools that behave predictably, not systems that generate new outputs every time.

We break down seven assumptions baked into nearly every AI policy in circulation, from treating AI as just another tool the city controls, to reducing governance to risk management, to mistaking disclosure for real transparency. The conclusion: these policies aren't poorly written. They're doing exactly what they were built to do, and that's the problem.

Source article: We have an AI Policy Problem

Find my work at ⁠micahgaudet.com⁠

Connect with me on ⁠LinkedIn⁠

Attitudes and Perceptions in AI Adoption Nov 23, 2025

Show notes

The AI revolution isn’t unfolding the way the headlines claim. New data from Pew, Gallup, OpenAI, and Anthropic reveals a far more contradictory story—one where rapid adoption collides with deep skepticism, and where AI’s most common uses are surprisingly ordinary.

In this episode, we break down five findings that challenge the popular narrative:

• Younger adults are the biggest AI skeptics.
• Most AI use is mundane knowledge work, not sci-fi invention.
• Users automate more but report higher satisfaction with collaborative AI.
• AI use is consistent across jobs but radically different across countries and states.
• Americans fear AI’s risks but overwhelmingly want more training—not more rules.

Taken together, these insights reveal a public that’s uncertain, conflicted, and still figuring out how to integrate AI into daily life. The real story isn’t about the technology’s pace of change—it's about how we choose to use it.

Find my work at www.civicinnovation.ai

Deepfakes in Disasters Nov 15, 2025

Show notes

In this episode, we explore how synthetic media—specifically deepfakes—are no longer just online hoaxes but are fast becoming a critical operational threat for emergency management. Drawing on Micah Gaudet’s piece “Deepfakes in Disasters: The New Operational Threat for Emergency Management,” we examine how disasters rely on speed, trust, and clear communication—and how manipulated audio-visual content can disrupt those foundations. linkedin.com

We discuss:

Why deepfakes aren’t just about identity theft or memes—they may undermine the first hours of crisis response.
How emergency operations built for clarity and coordination can be thrown off by credible but false imagery, false instructions, or mis-info timed to sow confusion.
What public-safety and local government leaders must do now: build detection awareness, craft communication resilience, and update incident playbooks to include not just physical threats but synthetic media threats.
Practical steps for cities and agencies to harden operations against deepfakes and regain control of trust when every second counts.

If you’re responsible for public safety, emergency planning, city management, or simply want to understand how tech-driven risks are evolving faster than policy and training—this episode offers a clear wake-up call and a path toward preparedness.

Follow my work at www.civicinnovation.ai

The AI Threat No One Is Talking About Nov 15, 2025

Show notes

This episode breaks down a risk almost no one is talking about: what happens when AI works. Not when it fails—when it quietly reshapes judgment, accountability, and trust without anyone noticing. Micah Gaudet explains why AI’s “uninsured risk” isn’t bias or bad outputs, but the slow erosion of human oversight as organizations automate what used to depend on presence, discretion, and relationships. If you work in government, policy, or AI strategy, this will reframe how you think about risk—and what real stewardship looks like in an AI-driven world.

Follow my work at www.civicinnovation.ai

How do We Spend the Time AI Saves? Nov 14, 2025

Show notes

AI promises to save us time—automating emails, summarizing meetings, and taking routine work off our plates. But history shows the opposite happens: every hour saved gets filled again. Productivity goes up, but rest, focus, and creativity go down.

In this episode, we explore the paradox at the heart of modern work: why efficiency rarely leads to freedom. We unpack how our attention spans collapsed to 47 seconds, why busyness became a badge of honor, and how AI might quietly deepen the problem it’s supposed to solve.

You’ll hear why the real challenge isn’t artificial intelligence—it’s our default settings. Because unless leaders protect the margin AI creates, the time it saves will vanish into noise, notifications, and shallow work.

The question isn’t whether AI will make us more efficient. It’s whether we’ll use that efficiency with wisdom.

www.civicinnovation.ai

When AI Becomes Political Oct 16, 2025

Show notes

Artificial Intelligence is reshaping how governments work—from automating services to predicting needs before they arise. But along with the promise of greater efficiency and accountability come real risks: bias, loss of transparency, and erosion of public trust.

In this episode, we explore how cities and nations are using AI to improve decision-making, service delivery, and oversight—and what happens when they get it wrong.

We’ll unpack what it means to build an “AI-ready culture” inside government, why trust—not technology—is the hardest part of adoption, and how new frameworks like the EU AI Act are reshaping the rules of the game.

Join us as we look at the real-world tension between efficiency and ethics, and ask: can AI make government more human—or just more mechanical?

Show Notes:

Stop Selling Sci-Fi - Why Gov AI should be Boring Oct 16, 2025

Show notes

Public sector AI is often sold like a sci-fi movie—glowing data lines, futuristic skylines, robotic hands. But the real barriers aren’t cinematic. They’re structural.

In this episode, we unpack why AI in government still feels distant and unrelatable. The problem isn’t the tech—it’s the story we tell and the systems we ignore. We explore why the real goal isn’t to master AI, but to use it quietly and effectively to serve people better.

From OpenAI’s simple, human-centered ads to the hard reality of budgets, procurement, and job descriptions, we trace how to make AI belong in daily work. The takeaway: innovation doesn’t mean flashier tech. It means making AI as normal—and as boring—as a spreadsheet.

Show Notes:

The Collapse of Clicks: How AI Changes Public Information Sep 11, 2025

Show notes

AI-generated summaries at the top of search results are quietly reshaping how residents get their information. Instead of clicking through to articles—or even city websites—many stop at the AI overview. That behavior shift is collapsing referral traffic for publishers and raising new challenges for local governments that depend on clarity, accuracy, and trust.

In this episode, we focus on the implications for cities and towns:

Why declining clicks to local media outlets matter for civic accountability.
How AI overviews could misrepresent policies, notices, or emergency information.
What residents’ demand for speed and simplicity means for government communication.
Strategies local governments can adopt—like machine-readable data, direct channels, and trust-building—to stay visible and credible in an AI-first search world.
Why the real issue isn’t just AI, but changing resident expectations shaped by years of frustration with cluttered news sites.

The zero-click future is already here. Local governments must decide whether they will be left out of the conversation—or become the trusted answer residents find first.

Show Notes:

GPT-5 for Government Leaders Sep 03, 2025

Show notes

OpenAI’s GPT-5 is here—and it changes more than you think. While the leap from GPT-4 to GPT-5 may look incremental, the impacts on government workflows, prompting strategies, and governance are anything but small. This episode unpacks what GPT-5 actually brings: router-style architecture, verbosity and reasoning controls, free-form function calling, improved memory, reduced hallucinations, and more proactive “it just does stuff” behavior.

We also examine why these changes matter for local government. From disrupted automations to the need for explicit prompting strategies, GPT-5 requires a shift in mindset: treat it as infrastructure, not novelty. Learn practical ways to manage governance, set boundaries, and unlock higher-quality outputs by using verbosity settings, reasoning levels, rubrics, and the “router mindset.”

If you work in or with government, this episode gives you the briefing you need to understand both the risks and opportunities of GPT-5.

Show Notes:

SGR Wednesday Webinar: Unlocking the Power of GPT-5

AI in Government LinkedIn Newsletter: Prompting GPT-5 in Local Government: Length, Voice, and the Router Mindset

Learn more about Micah's work