Click any tag below to further narrow down your results
+ ai
(3)
+ technology
(2)
+ financial engineering
(1)
+ inflation
(1)
+ expertise
(1)
+ infrastructure
(1)
+ biosecurity
(1)
+ far uvc lamps
(1)
+ ai hierarchy
(1)
+ railroad analogy
(1)
+ software market dominance
(1)
+ business technology
(1)
+ efficiency paradox
(1)
+ spreadsheet history
(1)
+ excel
(1)
Links
This piece argues that falling prices for digital goods, driven by cheap AI inference and open-weight models, won't pull overall inflation down much, because services like nursing, electrical labor, and housing dominate the cost of living. It projects that the Fed holds rates at 3.75% and stops cutting at 3.00% even in a credit crunch.
- Digital goods are about 5% of the consumer CPI basket and could deflate 4–5% a year by 2030 as inference gets cheaper.
- The other 95% physical services basket, including hospital nursing ratios, union electrical raises, and housing, stays near +3% annually.
- Headline inflation floors around 2.5%, so the Fed stays at 3.75% in the baseline and stops cutting at 3.00% even in a credit crunch.
The author, a former theoretical physicist, argues that superintelligence won't cause the explosive economic change rationalists predict, because physical, regulatory, and informational bottlenecks cap how much faster the overall world can move. He uses Amdahl's Law and a widget-factory example to show that speeding up one step rarely speeds up the whole process by much.
- Amdahl's Law: if a process has several sequential bottlenecks, infinitely speeding up one still leaves a constant-factor overall speedup, so exponential gains require repeatedly removing the current bottleneck.
- In the factory example, even with omniscient AI agents working 24/7, a month-long fix only drops to about a week because of shipping times, lab processes, and installing physical hardware.
- Some bits of information are very expensive to acquire (like long-term side-effect data), and regulation and political will are slow speed limits that intelligence doesn't remove.
The author argues that making software cheap is likely to mean much more software, as Jevons predicted for coal in 1865, but that whether the people who build it gain depends on what the new tool does to their specific job. Historical cases like farming, light, bank tellers and telephone operators show the paradox holds for output but can fail for workers.
- Jevons argued in 1865 that efficient engines would increase coal use, and that cheaper products would widen employment, which the coal boom bore out for about fifty years before output collapsed.
- The price elasticity of demand decides the outcome: food (elasticity roughly 0.27 to 0.81) saw the savings leave the industry, while light behaved as if elastic and use rose thousands of times over.
- Bank tellers survived ATMs because branches multiplied, but the job shrank from 600,500 in 2008 to 339,200 in 2025 once mobile banking removed the need to visit, and telephone operators lost 50 to 80% of their jobs almost immediately after mechanical switching.
- Software output already looks like coal in 1865: Google's token processing rose about 330 times in two years, and new repositories and app submissions climbed sharply.
This DeepMind essay argues that smarter AI alone won't reshape the economy much; the big shift comes when AI agents can coordinate with each other the way effective human teams do. It maps capability thresholds for competence and coordination and tracks how market structure, employment, and wages would change across them.
- Most AI benchmarks test individual agents in isolation, but economic outcomes come from organizations, where a well-structured team can beat a stronger solo performer.
- In a 100-person quiz across six domains, people were jagged: skill in one area barely predicted skill in another. Pairing complementary specialists outperformed random pairs by 32%.
- AI agents can be copied on demand at the cost of compute, so once a model matches human capability, human wages become tied to compute prices. Models also tend to be jagged in similar ways, so adding copies doesn't add breadth the way adding a complementary human does.
The author argues that fears of mass AI unemployment rest on the "lump of labor fallacy"—the mistaken belief there's a fixed number of jobs. History shows that as technology automates old work, humans invent new desires and new jobs to fill, a pattern unlikely to break now despite AI's capabilities.
- Unemployment rates haven't risen over 80 years despite continuous automation since the Industrial Revolution, yet 71% of Americans expect AI to cause job losses in the next 20 years—a historical pattern reversal that needs stronger evidence than speculation.
- Humans have an infinite capacity to generate new wants once current ones are satisfied (the "hedonic treadmill"), which creates perpetual demand for new jobs; jobs like "professional screamer" and "sorority rush consultant" aren't signs of economic decadence but proof this mechanism still works.
- Current data contradicts AI displacement predictions: Indian IT companies' headcounts haven't fallen, radiologist salaries hit all-time highs, and over 329,000 bank tellers still work in the US despite ATMs and mobile banking supposedly making them obsolete.
The author argues that LLMs don't meet the criteria that make technologies truly "inevitable" — they can't match human quality at high levels, they only replace low-value work, and they don't deliver the overwhelming performance gains that force adoption. Just because a technology is powerful doesn't mean society has no real choice about using it.
- A technology becomes inevitable only when it massively outperforms alternatives, performs socially valuable work, and leaves no way to neutralize its impact without radical behavioral change. Firearms met all three criteria in medieval Europe; the magnetic compass didn't, and remained optional.
- LLMs can only replace the least valuable parts of knowledge work — mediocre code, superficial research, low-quality design — while human expertise remains essential for anything requiring real quality. The more important the work, the less useful LLMs become.
- Real-world LLM efficiency gains are modest (20-50% improvement in low-stakes work like dashboards) and shrink dramatically as you approach actual business bottlenecks. If LLMs were truly overpowering, we'd already see tech companies producing vastly more output with them.
Ezra Klein interviews Nvidia CEO Jensen Huang about artificial intelligence, focusing on Huang's perspective that AI alarmism has been overblown. Huang argues that Nvidia's GPU chips—originally designed for gaming—became the foundational hardware that made modern AI possible.
- Nvidia controls the infrastructure layer of AI: since 2023, 15 cents of every dollar in US stock market returns came from Nvidia stock, and the company is now valued at $5.4 trillion
- Huang contends that frontier AI labs (OpenAI, Anthropic, Google) aren't the only voices worth hearing, and that AI risk concerns have gone too far
- Nvidia's GPUs weren't designed for AI but turned out to be perfectly suited for the parallel computing required by deep learning, making modern AI possible rather than merely popular
Planet Money is an NPR podcast that explains economic concepts through narrative-driven stories rather than dry analysis. It treats economics like a conversation with a friend—accessible, surprising, and actually engaging.
- The show makes economics approachable by grounding abstract concepts in real stories and human situations
- It's built on the premise that learning about the economy doesn't have to feel like a lecture
Two economists argue that while AI capabilities are exploding, the jump to 10%+ annual GDP growth in the next 15 years is extremely unlikely—not because of the technology, but because of economic frictions like unautomatable jobs, scarce resources, and measurement gaps. They show the math: doubling wealth once in 15 years only requires 4.7% annual growth, which would already be massive.
- Thinking in wealth levels rather than growth rates reveals how extreme double-digit predictions actually are: 16.6% annual growth means being 100x richer in 30 years, not just "a bit better off"
- While standard growth models can theoretically produce explosive AI-driven growth by eliminating labor as a bottleneck, this requires five unrealistic assumptions to all hold simultaneously—including complete economy-wide automation and zero disruptions
- A more reasonable baseline for the next 10-15 years is 4-5% growth, and the authors have bet money that US per capita real GDP growth stays below 15% annually through 2033
Most questions about AI aren't actually technical questions—they're about philosophy, economics, psychology, and how society will transform. Technical AI experts are often unqualified to answer them, just as steam engine engineers weren't the ones who understood the Industrial Revolution's broader impacts.
- The AI revolution, like the Industrial Revolution, will transform institutions, culture, economics, and human psychology all at once, making it impossible to understand through narrow technical expertise alone
- Even questions directly about AI—like whether super-intelligent systems will be conscious, or whether "intelligence" is a single measurable property—depend on philosophy and theory of mind, not just how neural networks work
- The core assumption of modern expertise (that reality decomposes into independent domains) breaks down when one technology threatens to reshape everything simultaneously, requiring polymaths and philosophers rather than siloed specialists
Three essays won a contest on major AI challenges. First, Jassi Pannu outlines a $40–60 billion plan to end airborne disease transmission using far-UVC lamps and infrastructure. Second, Ege Erdil advises middle-power countries to boost growth through proven policies like strong property rights and low taxes. Third, Michael Li likens AI labs’ business model to Hong Kong’s MTR, suggesting labs buy up complementary assets to offset high CapEx.
- $40–60 billion over ten years on far-UVC lamps and passive infrastructure could cut seasonal flu deaths by 60% and reduce pandemic odds tenfold
- Middle-power countries without AI hardware/software can still win by sticking to boring proven policy: strong property rights, low capital taxes, open regulations, rather than drastic moves
- AI labs could offset massive compute/R&D costs by owning "complementary assets" around their core business, similar to how Hong Kong's MTR subsidizes rail with adjacent real estate
The article traces tech’s rise from cloud in 2016 to today, showing software firms now rival entire economies in market cap. It then draws parallels to 19th-century railroads, explores AI’s potential to reshape corporate hierarchies, notes stablecoins shifting toward payments, and examines plunging trust in mass media among younger generations.
- The ten largest public companies by market cap now exceed the combined GDP of the G7 (excluding the US), achieved within roughly a decade of cloud computing's rise.
- Railroads once commanded up to 63% market share and forced the invention of modern corporate hierarchy; some argue AI could similarly flatten management structures today.
- Stripping out speculative and treasury flows, stablecoins generated $350–550 billion in genuine payment transactions last year, with consumer usage growing fast.
- Trust in mass media has collapsed from 72% in 1975 to 28% in 2025, with young Americans increasingly relying on social platforms instead.
This article explores the profound impact of electronic spreadsheets, particularly Microsoft Excel, on American businesses and the economy. It traces the evolution from pre-spreadsheet management practices to the modern reliance on data-driven decision-making and financial engineering. The piece also touches on the implications for future technologies like artificial intelligence.
- About one-sixth of the world's population uses Excel, yet the tool gets little credit for how thoroughly it reshaped business.
- Spreadsheets shifted corporate focus from production toward numerical optimization and financial engineering.
- Before spreadsheets, managers tracked operations with columnar pads and typewritten memos, making real-time analysis of complex data essentially impossible.
- Dan Bricklin's classroom-inspired idea, developed with Bob Frankston at Software Arts, turned into the electronic spreadsheet that gave companies unprecedented speed and precision in analyzing their own operations.
TVs have dramatically decreased in price over the past 25 years, primarily due to advancements in liquid crystal display (LCD) technology and manufacturing efficiencies. Key factors include the scaling up of mother glass sheets, improved manufacturing processes, and higher yield rates, which have all contributed to the significant reduction in cost.
- LCD manufacturing shifted to ever-larger "mother glass" sheets, letting factories cut many more panels from each sheet and slashing per-unit cost.
- Manufacturing process improvements and higher yield rates (fewer defective panels per batch) drove costs down further.
- These combined efficiencies explain why TV prices have fallen dramatically over the past 25 years.
Making software development easier leads to an exponential increase in the amount of software created, rather than a decrease in the need for developers. As tools and abstractions reduce the cost of building software, previously unviable projects become feasible, shifting the focus from whether to build something to what should be built. This pattern reflects a consistent trend across technological advancements, indicating a growing demand for knowledge work.
- Lowering the cost of building software doesn't shrink developer demand—it expands the pool of projects worth building, increasing overall software output exponentially.
- The bottleneck shifts from "can we build this?" to "what should we build?" once technical barriers drop.
- This mirrors historical patterns from other technological efficiency gains, where easier production led to more consumption/creation rather than less labor demand.
- Points to sustained, growing demand for knowledge work rather than obsolescence as tools improve.