AI summary by TwinMind

AI Hype: “Billions of dollars will be incinerated” Business Analysts Warn

Sabine Hossenfelder

6m 52s

AI Optimist Predictions

  • Goldman Sachs: AI could raise global GDP by 7% in next 10 years
    • Adding ~$7 trillion value to world economy
  • McKinsey: Even more optimistic projections
    • Generative AI could add $2.6-$4.4 trillion to global economy annually
  • Center for Economic Policy Research conference survey
    • Most attending economists expected 4-6% GDP boost

Current AI Revenue Reality

  • Very few companies seeing revenue from AI services
  • OpenAI (ChatGPT): Making money, reportedly reached $2 billion
  • Anthropic (Claude): Expects ~$850 million by end of year
  • Meta/Microsoft: Hard to determine AI-specific revenue
  • Hardware companies profiting most:
    • Nvidia and TSMC producing specialized AI microchips
  • Driving startup boom with everyone creating AI applications
  • Likely bubble that will burst soon

Skeptic Arguments

Jim Covello - Goldman Sachs Global Equity Research Head

  • Upfront AI costs enormous, difficult to recoup investment
  • AI applications must solve extremely complex problems for appropriate ROI
  • AI infrastructure buildout estimated to cost over $1 trillion in next several years
  • Key question: "What $1 trillion problem will AI solve?"
  • If AI companies profitable, cheaper to stick with low-wage workers

Daron Acemoglu - MIT

  • Published paper re-evaluating AI impact on US economy
  • Impact will be far lower than expected for two reasons:
    1. Not all jobs that can be replaced by AI can be done cheaper by AI
    2. Even if AI can do job cheaper, won't be much cheaper
  • Estimates: ~20% of labor could be replaced, only 1/4 of that (~5% total) will be replaced
  • Savings will be ~27%
  • GDP impact: 1% at most (far below optimistic estimates)

David Cahn - Sequoia Capital

  • Published essay "AI $600 billion question"
  • Big gap between revenue expectations from AI infrastructure buildout vs actual revenue growth
  • Most AI startups: either much longer for revenue to materialize or never will
  • "Billions of dollars in capital almost certain to be incinerated"

Market Consolidation Prediction

  • AI requires long training time and high maintenance effort
  • Only few very big companies will survive long-term
  • Current major players: OpenAI, Meta, Google, maybe Anthropic
  • Rest will be applications built on top of these platforms
  • Elon Musk has chance to catch up
  • Governments advised to build own AI or risk dependence on tech giants

Sponsored Content

  • Brilliant.org course recommendation
  • Offers courses on neural networks, large language models
  • Interactive visualizations, Python scripts, videos
  • Free trial link: brilliant.org/sabina
  • 30 days free trial, 20% off annual premium subscription

TwinMind

Get summaries like this for your own meetings

TwinMind takes notes in the background during meetings, lectures, and calls, then writes the summary for you.

Free plan to get started

No bot joins your calls

Phone, desktop & Chrome apps

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome