How fast is AI?Explore the curve

How fast is AI?

You are here on the AI change curve.

For 150 years, US living standards rose ~2% a year, through electricity, the transistor, and the internet. AI may be the next of them. The question isn't whether it's powerful, but how fast that power turns into growth. The answer hides in a single idea: weak links.

Slow, real upside. Fast, fragile downside. Does AI finally bend the 2% line, or just hold it? Here's the evidence, the forecast, and the hardest objections, updated as the news arrives.

US income / person (history)World income / person≈2% / year trendBaselineFull automationIncomplete automation

Source: US and world history approximated from Maddison/BEA patterns; the world line grows slower early, then accelerates as the rest of the world catches up. Forward paths follow the published “Continuing the Past” growth scenarios. · Aghion, Jones & Jones (2019)

~100M
ChatGPT users
in ~2 months, fastest ever at the time
150 yrs
of ~2% growth
through every prior transformative tech
4.3% → 3.0%
computer share of GDP
falling since 2000, despite more chips
20+ yrs
self-driving diffusion
from “solved in 5” to still-rare

The puzzle

Transformative technologies, and still 2% a year.

Electricity, internal combustion, antibiotics, semiconductors, the internet: each wildly transformative, yet growth never strayed far from 2%. Each new technology kept 2% going for another 50 years as the previous one ran out of steam.

Real income / person ≈2% / year
ErasElectricity1882Mass production1908Transistor1947Microprocessor1971The Internet1991Generative AI2022

Source: US real income per person, 2025 dollars (approximated for display).

Computers are everywhere, but their price falls faster than their quantity rises. The weak links capture the value.

Source: Computers’ share of US GDP, approximated from the BEA/BLS value-added pattern.

The chart that flips the worry

Computers are the most plentiful thing in the economy, and their share of GDP has fallen by a third since 2000. Price falls faster than quantity rises. The plentiful thing gets cheap; the scarce thing, humans and the weak links they hold, captures the value.

When you worry AI will automate everything, remember: abundance drives down the price of what's abundant. Scarcity is where the returns go.

Where we are today

The concrete record: milestones, predictions, bottlenecks.

Not abstractions. Real dates, real numbers, and a scoreboard of who called it right. The pattern: adoption can be lightning-fast while economic transformation stays slow.

Milestone timeline

  1. Mar 2004Verified

    DARPA Grand Challenge: zero finishers

    Not a single autonomous vehicle completed the desert course. The starting gun for self-driving, and a reminder of how hard the physical world is.

  2. Oct 2005Verified

    Stanford's "Stanley" wins the DARPA Grand Challenge

    One year after zero finishers, Sebastian Thrun's team completed the 132-mile course. Twenty years later, robotaxis are still rare outside a few cities.

  3. Sep 2012Verified

    AlexNet ignites the deep-learning era

    A deep neural network crushed the ImageNet benchmark, kicking off the modern wave of AI capability gains.

  4. Mar 2016Verified

    AlphaGo defeats Lee Sedol

    DeepMind's system beat a top human Go player 4–1, years ahead of expert expectations for the game.

  5. Jun 2020Verified

    GPT-3 released as a developer API

    A 175B-parameter model showed broad few-shot ability, but reached developers, not a mass consumer audience.

  6. Nov 2020Verified

    AlphaFold cracks protein structure prediction

    At CASP14, AlphaFold2 reached near-experimental accuracy, transforming structural biology. The work earned a 2024 Nobel Prize in Chemistry.

  7. Nov 2022Verified

    ChatGPT launches

    OpenAI released a conversational interface over GPT-3.5. Adoption was almost immediate.

  8. Jan 2023Verified

    ChatGPT reaches ~100M monthly users in ~2 months

    The fastest-growing consumer app at the time. NOTE: this was ChatGPT (GPT-3.5), not GPT-3. Adoption can be blindingly fast even when economic transformation is slow.

    ~100M users in ~2 monthsReuters (Feb 2023)
  9. Mar 2023Verified

    GPT-4 released

    A large multimodal model with markedly stronger reasoning and coding, the workhorse behind the first wave of AI copilots.

  10. Aug 2024Verified

    Waymo scales paid robotaxi rides

    Driverless rides became a daily reality in San Francisco and Phoenix, yet remained rare nationally. Diffusion measured in decades, not years.

  11. Jun 2026Projection

    Today: you are here (mid-2026)

    Adoption is nearly universal; the economy-wide productivity jump is still small. Lightning-fast in a few lanes, slow across the rest, exactly the split the weak-link view predicts.

    How fast is AI?, present-day marker

Signals to watch, not yet verified

Forward-looking markers, kept out of the verified record. Treat each as a claim until a primary source confirms it.

  1. Nov 2025Claim

    A frontier model reportedly tops an engineering take-home exam

    A leading lab's multi-hour take-home hiring exam was reportedly completed by its newest model at a score higher than any human on record. Treat as a claim pending public confirmation.

    Reported, pending verification
  2. Dec 2025Claim

    Coding agents start holding context across multi-hour tasks

    The frontier shifts from quick answers to agents that run for hours, the first credible long-horizon autonomy in software.

    Unverified signal
  3. Feb 2026Claim

    Enterprises move AI agents into real workflows

    Pilots turn into production: support, coding, and back-office agents handle a slice of real volume while humans still own the judgment calls. Adoption races ahead of measured productivity.

    Unverified signal

Prediction scoreboard

Geoffrey HintonOver-predicted

"We should stop training radiologists", none will have jobs within five years.

By 2026 there are MORE radiologists than in 2016, and they are paid more. AI reads scans; humans handle the weak links: consults, hard cases, sign-off.

Lesson: weak linkssource

Elon MuskOver-predicted

A Tesla will drive itself coast-to-coast by 2017, with full self-driving roughly two years away.

A decade later, supervised "Full Self-Driving" still needs a human behind the wheel, and driverless robotaxis run in only a handful of cities. The physical world is full of weak links.

Lesson: diffusion takes decadessource

Almost everyoneUnder-predicted

A chatbot won't reach hundreds of millions of users quickly.

ChatGPT hit ~100M monthly users in ~2 months, the fastest consumer-app adoption on record at the time.

Lesson: adoption can be fast even when transformation is slowsource

AI-2027 / Situational AwarenessStill open

Explosive, economy-wide acceleration arrives within ~3–4 years.

Still open. The weak-link view expects real transformation measured in decades, with the downside risks arriving sooner.

Lesson: horizon is the cruxsource

Find the weak link, by domain

For each field: what AI has automated, and what remains the human bottleneck. Automating 75% of the tasks doesn't finish the job. The rest set the pace.

Software engineering

The first thing being automated, and still bottlenecked by judgment and accountability.

Automated

  • Autocomplete & boilerplate
  • Test generation
  • Bug discovery in mature code
  • Routine refactors

Weak link

  • System design & architecture
  • Ambiguous requirements
  • Integrating with messy production data
  • Owning the outcome when it breaks

Even if AI writes most code, integrating it into every business is a long, engineer-heavy process.

Radiology

Hinton's 2016 test case. AI got better at reading scans; the job grew anyway.

Automated

  • Scan triage
  • Cancer-detection assist
  • Measurement & flagging

Weak link

  • Surgical & treatment consults
  • The hardest, ambiguous scans
  • Liability and sign-off
  • Talking to patients and clinicians

More radiologists in 2026 than 2016, and better paid. Weak links win.

Driving

Seemingly simple, actually decades-long. The canonical slow-diffusion case.

Automated

  • Highway driving
  • Mapped-city autonomy (SF, Phoenix)
  • Sensing & lane control

Weak link

  • Long-tail edge cases
  • Unmapped regions
  • Adverse weather
  • Nationwide scale & trust

20+ years from "solved in 5" to still-rare. The physical world bottlenecks.

Early-childhood teaching

A 'someday' automation gated almost entirely by trust and safety, not capability.

Automated

  • Content delivery (potential)
  • Practice & feedback (potential)

Weak link

  • Trust & safety with children
  • Care and supervision
  • Parental acceptance
  • Liability

We could build a world-class teaching robot before we'd let it run a kindergarten unsupervised.

The forecast

Two calibrations, three scenarios, one slow explosion.

Calibrate the weak-link model to history and run it forward. Even the aggressive “Moore's Law everywhere” case takes ~30 years to fully play out. Growth explodes, but slowly, because you must automate every weak link first.

Baseline Full automation Incomplete automation

Source: Calibrated to the published growth scenarios (anchor points, interpolated).

1.04×
richer by 2050
Continuing-the-past baseline
1.38×
richer by 2100
…the explosion is slow
3.2×
richer by 2040
Moore’s-Law-everywhere case
~2050
explosion completes
even in the aggressive case

What it means for you

You're safe until about 2060. Probably.

Here's the uncomfortable, oddly reassuring logic. Even after the aggressive scenario's growth explosion runs its course around mid-century, the last things to automate are the weak links you already hold: judgment, accountability, presence, trust. Pick your role or your business and see how long those links keep you in the loop.

When does AI stop needing you?

Routine work goes first. What keeps you in the loop is the link AI is worst at: owning the call.

Needed until

~2046

That's about 20 more yearsof being the weak link AI can't replace.

AI already does~78%

of the routine work to close the books and sign off on the numbers.

What keeps you needed

  • Edge-case judgment
  • Audit accountability
  • Client trust

The routine ledger work goes first. Putting your name on the numbers, and answering for them, stays human longer than the spreadsheets do.

20302040205020602070

A thought experiment, not a forecast. The years are round anchors on the aggressive scenario: the macro growth explosion completes around 2050, and these human weak-links, accountability, care, the human handshake, are the last to fall, stretching toward 2060. Some may never fully fall, which is the “if ever” in the question.

Note:a “safe until” year is when the weak links start to bite, not a cliff where the role vanishes overnight. Automation erodes tasks gradually, and can even grow a field: AI made radiology better and there are more radiologists now, not fewer. These are round, illustrative anchors from the weak-link logic, not forecasts. Some links, accountability, care, a human hand, may never fully fall.

These dates will move.

I track the AI change curve as the news arrives: capabilities, adoption, and the weak links that still won't budge. Follow along on LinkedIn.

Subscribe on LinkedIn

The fast downside

Slow to improve. Fast to break.

The same weak-link structure that delays the benefits makes the risks arrive sooner. Strengthening a chain is slow, link by link. Breaking one link is instant.

The bad actor with a jailbroken oracle

Near-term

Frontier models keep getting jailbroken, often within days of release. Hand a bad actor a model that can do what the smartest humans can, and ask it to design a pathogen more lethal than Ebola with a three-month latent period. We survived nuclear weapons because only a handful of people held the button. What happens when billions do?

Open-source bug-hunters turned on the grid

1–3 years

Models are already finding bugs humans missed in decades-old, battle-tested software. It is not hard to imagine a capable open-source version in many more hands within a year or two. How sure are we no one points it at the electric grid, the banking system, or a bio lab? Not an existential problem, but a plausible one, soon.

Retaining power over smarter entities

Speculative

When more advanced species have met less advanced ones in history, it hasn't gone well for the less advanced. Stuart Russell's question is the uncomfortable one: how do we retain power over entities more capable than us, forever?

Slow to improve, fast to fail

Structural

The same weak-link structure that makes the upside slow makes the downside fast. A chain takes enormous effort to strengthen link by link, but breaking a single link destroys the value instantly. The Space Shuttle Challenger was lost to a $25 O-ring. That asymmetry is the core warning.

Strongest objections

The best arguments against this view.

A reference earns trust by steelmanning its critics. Here are the strongest arguments against this view, in their sharpest form, with honest responses.

This is the crux, and it lands. The forecast's slowness comes from calibrating to a past where the hard tasks never moved much. If AI keeps climbing at exactly the work we've labeled “human,” the gradual story collapses and the upside arrives far sooner. The honest position: the speed depends entirely on how strong those weak links really are, which is the one number we're least sure of.

The gradualism is an assumption, not a law. The cognitive weak links may not hold.

So who is scarce?

In a world of abundance, the weak links pay.

Scarcity is where returns go. As AI makes cognition abundant, value flows to whatever stays scarce, judgment, accountability, the human-held links, and to whoever owns the capital. Here's the practical read.

Be the manager

Someone consults the AIs and makes the final call. That judgment becomes scarce and valuable.

Own the capital

If labor's share falls, owning a slice of the S&P 500 means you share in the capital's gains.

Redistribute deliberately

Abundance makes good outcomes possible, not automatic. Tax-and-transfer is a choice, not a guarantee.

Find meaning anyway

Retirees in a world of abundance still find purpose. Pottery, friends, and learning the latest model.

For change leaders: the same story at company scale

The “weak links” that bottleneck a whole economy are the same shape as the weak links inside your organization. The macro thesis and the org-change literature meet at the metaphor.

Macro storyOrganizational analog
Normal tech vs FOOMTheory of change vs theory of changing
Continuation vs breakFirst-order vs second-order change
Weak links bottleneck gainsSensemaking, identity, coordination cost, OCC
Diffusion takes decadesTsoukas & Chia: organizational becoming
Fragile downsidePerforming → reinforcing → breakdown → reflecting
Scarcity earns returnsResistance as diagnostic, not obstacle

The practical move: sequence your automation across the process chain and treat employee resistance as the instrument that locates your weakest link, not as friction to overcome.

The library

The research behind the big claims.

The big claims here trace back to published work, and where a number is our own estimate, digitization, or projection, the page says so. The weak-link view of growth draws on Charles I. Jonesand the wider literature; Daron Acemoglu's estimates anchor the skeptical end.

Economic growth

Weak links & fragility

Labor & distribution

  • Karabarbounis & Neiman2014
    The Global Decline of the Labor Share

    Context for the capital-vs-labor split that the scenarios track to 100% / 0%.

  • Brynjolfsson, Rock & Syverson2021
    The Productivity J-Curve

    Why measured productivity lags transformative tech: adoption races up the S-curve while output sits in the J-curve trough.

Risk & the fast-takeoff case

  • Dario Amodei2024
    Machines of Loving Grace

    The optimistic 'country of geniuses in a datacenter' vision, the aggressive end of the forecast dial.

  • Leopold Aschenbrenner2024
    Situational Awareness

    The fast-takeoff case (explosive growth by ~2027) that the weak-link view pushes back on.

Organizational change