More The Reasoning Show episodes

AIs Netflix Moment: What Early Cloud Teaches Us About the Next Phase thumbnail

AIs Netflix Moment: What Early Cloud Teaches Us About the Next Phase

Published 11 Nov 2026

Duration: 00:21:40

Enterprise AI is shifting from experimental token maxing to ROI-driven adoption, creating a divide between users who leverage AI for complex tasks and those who do not.

Episode Description

Brian and Brandon discuss what happens to enterprise AI if the market corrects and the era of "move fast and break things" gives way to hard questions...

Overview

The discussion centers on what happens to enterprise AI adoption if the market experiences a correction or "bubble burst." The speakers explore the disconnect between those who have used AI to complete tasks that previously felt impossible and those who haven't experienced that shift. They draw parallels to the early cloud era, comparing the current moment to when Netflix demonstrated what was possible while most companies couldn't imagine reaching that level.

The conversation traces enterprise AI adoption through several phases: initial curiosity, corporate policy responses (often just purchasing Microsoft Copilot), shadow IT concerns, and the "token maxing" period of unlimited access. The speakers argue the next phase will force organizations to confront ROI questions, similar to how cloud adoption required fundamental changes in processes and organizational structure.

A key distinction is drawn between those who have "touched the superpower" of AI - completing a previously impossible task rapidly - and those who see it as just another technology. The speakers note this experience gap explains why some organizations pull back while others double down, and they suggest the coming period will be "bumpy" as market dynamics shift.

What If

  • What if you replaced your "yeah, but" reflex with a "what if" prompt and asked an AI agent to build your entire weekly reporting pipeline?

    • Move: Take your existing weekly reports or spreadsheets, prompt an agent to automate the full workflow, then run a beta with one colleague whose feedback emails you paste back into the agent for fixes.
    • Why Now?: The speakers describe a compressed timeline where "things could go a million miles an hour with AI", and a market correction may make it harder to justify new tooling later.
    • Expected Upside: You could potentially cut report generation from hours to minutes, and the beta feedback loop (fixes in "20, 30 minutes" per the source) means iteration is fast. This is a proposed workflow, not a guaranteed outcome.
  • What if you treated your next software project as a "vibe coding" engagement where the AI is "the best software developer you've ever worked with"?

    • Move: Define the goal in plain language, let the agent build the full application (not a tiny component), then fire up a separate security-audit agent on the output before any production deployment.
    • Why Now?: The speakers argue "no one actually sits down even in production settings and writes code", and the production gap (corner cases, security) is the real bottleneck, not the initial build.
    • Expected Upside: You could ship a working MVP faster and get a "professional level" security audit that "I would have never gotten... on the software I was building". This is a proposed approach; actual results depend on the specific project.
  • What if you identified the one task in your business you've "never really got around to" because it was "so hard" or "too long," and completed it end-to-end with AI this week?

    • Move: Pick a deferred task (a document summary, a small app, a data cleanup), give the agent the full goal, and iterate through feedback until it's done - rather than boxing AI into "one little thing."
    • Why Now?: The speakers frame this as the dividing line between those who've "touched the superpower" and those who see AI as "yet another technology". A market correction may pressure organizations to show concrete ROI.
    • Expected Upside: Completing one previously impossible task "incredibly rapidly" gives you a concrete story to demonstrate value, which the speakers say will make leaders "pull you out" for bigger work. This is a proposed benefit, not a source fact.

Takeaway

  • Replace "yeah, but" with "what if" prompts: before committing to any AI workflow, ask the agent "What if I asked you to do X, Y, Z?" and evaluate the output before building anything.

  • Stop boxing AI into single tasks. Give the agent the full goal for a project (e.g., "build the entire new website") and iterate through feedback rather than asking it to do "one little thing".

  • Run a two-agent workflow for production software: use one agent to build and fix issues from beta feedback, then fire up a separate security-audit agent to scan the output before any deployment.

  • Pick one deferred task you've avoided because it was "so hard" or "too long" and complete it end-to-end with AI this week to build the "superpower" experience that differentiates you from those who see AI as "yet another technology".

  • When applying AI to your business, target areas where you can "take feedback quickly and put it in places that it can succeed" rather than forcing it onto processes full of "rocks and salt and gears grinding".

Recent Episodes of The Reasoning Show

23 Sept 2026 How Do Regulated Enterprises Deploy Private, Secure AI?

"Enterprises prioritize secure, sovereign AI solutions to control data, avoid vendor lock-in, and balance cost, performance, and compliance while navigating AI agents, customization, and future adoption challenges."

9 Sept 2026 How Open-Source is Reshaping the AI Infrastructure Stack

"Open-source technologies like Kubernetes and Kubeflow are reshaping AI infrastructure, addressing challenges in adoption, emphasizing reproducibility and transparency, and exploring future trends like edge computing and hybrid deployments."

29 Jul 2026 How AI Stacks are rewriting the Rules of Business

"AI's shift from models to full software stacks is reshaping business operations, blending data, processes, and tacit knowledge into intelligent systems, while addressing challenges like scalability, governance, and economic sustainability."

22 Jul 2026 AI's Impact on Trust and Brand

"AI is transforming branding and marketing by enhancing efficiency but requires governance to mitigate risks like inconsistent brand representation, unstructured data challenges, and cost concerns, demanding strategic alignment with business goals."

More The Reasoning Show episodes