← All topics

AI & technology

AI hype, ethics & governance

37 indexed claims · 13 guests · peaked 2025-Q3

AI hype, ethics & governance

Overview

These podcast conversations map a field in the middle of a reckoning. Forty-nine claims from twenty voices across seven quarters chart the distance between what AI is actually good for and what the industry has been selling. The dominant tensions run in three directions: between AI's raw capability and the governance infrastructure needed to contain it; between the efficiency gains AI offers and the cognitive atrophy that follows when critical thinking is quietly outsourced; and between genuine consumer trust — hard-won through transparency and real control — and the hype-cycle framing that mistakes investment signaling for product value. At the center sits a question of professional identity: researchers who understand where AI breaks will displace those who resist it entirely, while the slower work of building trust norms, disclosure frameworks, and data-rights regulation struggles to keep pace with tools that move faster than law.

From the corpus

  • The most dangerous aspect of AI products is that they almost never tell the user when they are bad at a task or suggest an alternative tool — unlike a professional who would redirect you to the right expert, AI produces confident-looking output regardless of whether it can actually do the thing asked. — Rand Fishkin, Ep. 19 · cl-rand-fishkin-018
  • AI is designed to be a polite assistant that mirrors users' biases back at them in more intelligent-sounding form, making it a tool of intellectual comfort rather than courage; over-reliance on it for the wrong tasks dulls the critical-thinking muscles researchers most need to sharpen. — Sarah Montgomery, Ep. 22 · cl-sarah-montgomery-015
  • Rolling out agentic AI marketing workflows at scale without a validated, real-time representation of the customer is genuinely dangerous — the same automation power that enables personalization can ship the wrong message to the wrong segment at enormous scale before anyone notices. — Kate O'Keeffe, Ep. 55 · cl-kate-okeeffe-016
  • There is no ethical way to use AI because the foundational models were built by taking content without permission or compensation from creators; while a legal 'fair use' argument was accepted under political pressure, it does not meet the moral or dictionary definition of fair use. — Rand Fishkin, Ep. 19 · cl-rand-fishkin-013
  • SparkToro's audience data shows ChatGPT ranks seventh among search and AI tools by actual usage — behind Amazon, Reddit, YouTube, Google, and Bing — which means brands investing heavily in AI-channel presence are likely misallocating against media-hype rather than audience reality. — Rand Fishkin, Ep. 19 · cl-rand-fishkin-019
  • Newcomers entering consumer insights today should make curiosity, hands-on tool experimentation, and LLM literacy their daily practice — knowing both the powers and the limitations of AI tools — while learning the proven methodologies and rules of thumb from those who came before. — Lev Mazin, Ep. 24 · cl-lev-solo-017
  • Describing a product as 'AI-first' or 'AI-powered' in 2024 is as meaningless as describing a website as 'code-powered' in 2001 — the label is a VC-facing signal, not a customer-relevant description, and companies using it are almost exclusively doing so to attract investment. — Rand Fishkin, Ep. 19 · cl-rand-fishkin-015
  • Older adults actually read privacy statements before adopting technology — checking what data is collected, who the third parties are, and what will happen with their information; brands must signal on privacy because older adults pause where younger users click through. — Brittne Kakulla, Ep. 45 · cl-brittne-010

Contributing guests

Episodes in AI hype, ethics & governance

See AI & technology mapped across all quarters →