Skip to main content
The Knowledge Project

Why Everyone Is Wrong About AI (Including You) | Benedict Evans

73 min episode · 2 min read
·

Episode

73 min

Read time

2 min

Topics

Fundraising & VC, Artificial Intelligence, Product & Tech Trends

AI-Generated Summary

Key Takeaways

  • AI Adoption Reality: Survey data shows only 10% of people use AI daily, 15-20% weekly, with another 20-30% trying it monthly. Many look at ChatGPT and don't understand how to use it, similar to early spreadsheet adoption where value wasn't immediately obvious to all users.
  • Model Commoditization: Double-blind tests of prompts across Grok, Claude, Gemini, Mistral, and DeepSeek would likely be indistinguishable to most users. Models themselves lack differentiation, yet ChatGPT dominates usage through brand recognition and distribution, not superior technology, raising questions about sustainable competitive advantages.
  • Google's Reset Risk: The primary threat to Google isn't superior AI search, but a moment of discontinuity where users reconsider defaults and reset their behavior patterns. This creates openings for competitors even if Google maintains technical superiority, similar to how platform shifts historically disadvantaged incumbents.
  • Quantitative Analysis Limitation: AI currently has zero value for quantitative work requiring precise accuracy because it produces results that are roughly right but wrong dozens of times per page. It excels at qualitative tasks like drafting, brainstorming, and image generation where approximate correctness is acceptable and can be edited.
  • Regulation Trade-offs: Treating AI like nuclear weapons with tight controls, as the EU approach does, creates explicit policy trade-offs. Making it hard to build models and start companies will slow innovation, similar to how restrictive housing policy makes houses expensive—you can choose that outcome but cannot complain about the consequences.

What It Covers

Benedict Evans analyzes AI as a platform shift comparable to the iPhone, not a revolutionary transformation. He examines adoption patterns, incumbent advantages, competitive dynamics among tech giants, and why most people still don't use AI regularly.

Key Questions Answered

  • AI Adoption Reality: Survey data shows only 10% of people use AI daily, 15-20% weekly, with another 20-30% trying it monthly. Many look at ChatGPT and don't understand how to use it, similar to early spreadsheet adoption where value wasn't immediately obvious to all users.
  • Model Commoditization: Double-blind tests of prompts across Grok, Claude, Gemini, Mistral, and DeepSeek would likely be indistinguishable to most users. Models themselves lack differentiation, yet ChatGPT dominates usage through brand recognition and distribution, not superior technology, raising questions about sustainable competitive advantages.
  • Google's Reset Risk: The primary threat to Google isn't superior AI search, but a moment of discontinuity where users reconsider defaults and reset their behavior patterns. This creates openings for competitors even if Google maintains technical superiority, similar to how platform shifts historically disadvantaged incumbents.
  • Quantitative Analysis Limitation: AI currently has zero value for quantitative work requiring precise accuracy because it produces results that are roughly right but wrong dozens of times per page. It excels at qualitative tasks like drafting, brainstorming, and image generation where approximate correctness is acceptable and can be edited.
  • Regulation Trade-offs: Treating AI like nuclear weapons with tight controls, as the EU approach does, creates explicit policy trade-offs. Making it hard to build models and start companies will slow innovation, similar to how restrictive housing policy makes houses expensive—you can choose that outcome but cannot complain about the consequences.

Notable Moment

Evans reveals he doesn't reflexively use AI despite being a technology analyst, lacking use cases for brainstorming, summarization, or code generation. His work requires original insight beyond what AI would produce, using the test: if ChatGPT would say it, he won't publish it.

Know someone who'd find this useful?

Episode Transcript

It seems to me right now, you could do, like, a double blind test of the same prompt given to Grock, Claude, Gemini, Mistral, Deepsea. I bet most people wouldn't be able to tell which is which. Benedict Evans is a technology analyst known for his insightful takes on platform shifts in the tech industry. He sees AI differently than others. He spent decades spotting patterns others miss and dives into how people really use AI. Why is it that somebody looks at this and gets it and goes back every week, but only every week? The very high level threat to Google is that you have this moment of discontinuity in which everybody resets their priors that we consider set of faults. And so it's no longer just the default that you go and use Google. There's this sort of question for Apple around, does this not actually change the experience of what the smartphone is, what the ecosystem is? Does it end up kind of getting Microsofted in the sense that I wanna start with your most controversial take on AI. It's funny. My I suppose my take on AI, controversial take on AI, rather like my controversial take on, crypto as being a centrist in that seems to me very clear this is, like, the biggest thing since the iPhone. But I also think it's only the biggest thing since the iPhone. And there's a bunch of people who think, no. It's much more than that. It's, minimum. It's more like computing. And then you've got people going around and saying, no. This is more like, you know, the electricity or the industrial revolution or, you know, transhumanism or something. My sort of base case is to say, this is kind of another platform shift, and all the new stuff will be built around this for the next ten or fifteen years. And then there'll be something else. And so the impact on employment will be kind of like the impact on employment from the other platform shifts and the impact on the economy and productivity and intellectual property. And there'll be there'll be a whole bunch of different weird new questions just like there were a bunch of different weird new questions before. And then in ten years time, it'll just be software. Put this in historical context for us with other platform shifts. Everybody's saying this time is different, which everybody does that each platform shift out of the margin. What's the same? Well, that's the there's a there's a famous book about, financial bubbles called This Time is Different. Because people always say this time is different, and it always is. Like, the .com bubble was different to, like, the late eighties and the Japanese financial bubble was different to, you know, pick any other bubble you want. They're always different. But that doesn't mean they're not a bubble. And the same thing here, I have a diagram I use a lot from, 1995. …

Get the full transcript (15,091 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all The Knowledge Project transcripts →

You just read a 3-minute summary of a 70-minute episode.

Get The Knowledge Project summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

Books, tools, and gear mentioned in this episode

SignalCast may earn commission on purchases via these links.

Tools

  • by OpenAI

    Many look at ChatGPT and don't understand how to use it, similar to early spreadsheet adoption where value wasn't immediately obvious to all users.
  • Double-blind tests of prompts across Grok, Claude, Gemini, Mistral, and DeepSeek would likely be indistinguishable to most users.
  • Double-blind tests of prompts across Grok, Claude, Gemini, Mistral, and DeepSeek would likely be indistinguishable to most users.
  • Double-blind tests of prompts across Grok, Claude, Gemini, Mistral, and DeepSeek would likely be indistinguishable to most users.
  • by Anthropic

    Double-blind tests of prompts across Grok, Claude, Gemini, Mistral, and DeepSeek would likely be indistinguishable to most users.
  • by Google

    Double-blind tests of prompts across Grok, Claude, Gemini, Mistral, and DeepSeek would likely be indistinguishable to most users.

More from The Knowledge Project

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best Business Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's AI & Machine Learning Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into The Knowledge Project.

Every Monday, we deliver AI summaries of the latest episodes from The Knowledge Project and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime