Nov 3, 2026 · 27 min
Technology Made Gadgets Cheap, Not Life’s Essentials
AI Lurches Toward Recursive Self-Improvement
The episode frames today’s economic frustration as a consequence of uneven technological progress, with cheaper goods alongside rising housing and healthcare costs.
- 1Consumer goods and electronics became dramatically cheaper over the past 25 years.
- 2Housing, healthcare, and other essential expenses rose even as technology lowered prices elsewhere.
- 3This uneven progress helped fuel economic frustration and the rise of populist politics in the United States.
Don't miss
The episode’s key turn is the contrast between dramatically cheaper electronics and increasingly expensive housing and healthcare.
The brief
The episode previews the second installment of How We Got Here, which examines how technological progress and globalization reshaped daily costs over the past 25 years.
Consumer goods and electronics became dramatically cheaper, creating visible gains in affordability and access for many households.
That progress was uneven: housing, healthcare, and other essentials grew more expensive, widening the gap between cheaper products and costlier lives.
The episode connects this economic divide to mounting frustration and the rise of populist politics in the United States.
Its central question is not whether technology lowered prices, but why it made the wrong things cheap while essential security became harder to afford.
What was said on this episode
13 statements · 7 positive · 4 negative · 1 mixed · 1 neutral
AI models are approaching the ability to improve themselves.
“AI models are getting closer to being able to improve themselves.”
Listen at 2:26
Consumer personal agents may become provider cost centers requiring alternative monetization.
“since consumers aren't willing to pay for this category of product, they're just going to be a cost center for the people offering them, and they're going to have to be made up for in other ways.”
Listen at 6:21
Natural AI pricing could reduce spending and invalidate current company valuations.
“people don't value or can't afford to value the intelligence at that natural price enough to continue spending in the way that they have been, and thus it means that all the valuations of all these companies are off.”
Listen at 8:38
A broad shift to API pricing would not substantially change AI usage.
“I don't particularly think that the shift to API pricing for everyone actually changes all that much.”
Listen at 9:40
High-value AI users will pay usage costs as an ordinary business expense.
“The people who are getting enough value out of whatever plan they're on, whether it's $20 or $200, that they're willing to actually pay for the intelligence they consume and just treat it as a cost of doing business.”
Listen at 10:01
A large share of AI revenue comes from power users paying through APIs.
“a huge amount of the revenue on an actual percentage basis is coming from power users who are already paying via the API.”
Listen at 10:50
Measuring a capability as an RL target can improve it and reduce its cost.
“once a capability is measured and becomes a target for reinforcement learning, it dramatically improves and becomes more cost-effective.”
Listen at 17:38
GPT-6 Astra improved QUEN scores substantially by editing 0.016% of weights.
“Astra changed only 0.016% of the weights, and yet was able to improve QUEN's scores on math benchmarks by 6x and instruction following by 3x.”
Listen at 18:49
The MIT result demonstrates a clear mechanism for recursive self-improvement.
“this was clear evidence that recursive self-improvement is not only a possibility in the future, but now has a clear mechanism that's understood and has been demonstrated in an academic paper.”
Listen at 22:29
Major capability improvements have previously driven substantially greater AI adoption.
“The evidence that we have for that is that true step changes have unlocked new capabilities that have led to massively more adoption in the past.”
Listen at 24:42
Enterprises currently consume little AI intelligence despite sufficient model capability.
“They are, at the moment, consuming a vanishingly small portion of what they could, and it's not because the model capabilities aren't high enough.”
Listen at 25:14
Low enterprise AI adoption is caused by organizational inertia and difficult implementation, not model limits.
“it's not because the model capabilities aren't high enough. It's because inertia is real, corporate systems are lumbering, and adoption is hard.”
Listen at 25:19
Improved models do not necessarily translate directly into increased revenue.
“I'm just not sure model improvement has quite such a straight line to increased revenue as we might intuitively think.”
Listen at 25:42
Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.