all problems
progress & abundanceEducation & pedagogyhumans affected:high· updated 2026-06-09

Pedagogy at scale

Close Bloom's 2-sigma gap, every child deserves a personal tutor, and AI may finally make it tractable.

The scale of it

70%worsening

share of 10-year-olds in low/middle-income countries who cannot read a simple text

The capital on it

$3B/yr↘ fallingunderallocated · 0.02× fair share

Edtech venture funding worldwide. Excludes public education budgets (~$5T/yr, overwhelmingly status-quo delivery). Post-2021 crash: the boom peaked at ~$21B and innovation capital has fled.

20212024

source: HolonIQ — global edtech venture funding · confidence med · estimate, improvable by PR

The prize at the limit

$300Bin-the-limit market cap, if the team executes perfectly

A company that delivers a genuine tutor-quality education to anyone, in any language, for near-zero marginal cost becomes the default layer for human capital formation worldwide — the closest thing to compounding the entire species.

comparable: the "Google of learning" — a new education default · confidence low · a ceiling, not a forecast

The trade: demand is high, only $3B/yr of capital is flowing (0.02× its fair share), and the prize at the limit is $300B. This is a Request for Startups and a Request for Investors at once.

Whitepaper · v0.1 · open to refutation

The summary lives here. The full whitepaper walks through the four-axis ranking, existing alternatives, proposed direction, cost & scale, and suggested investors — in the spirit of Hyperloop Alpha.

Quantity · humans affected

1.5Bchildren undereducated

source: UNESCO Global Education Monitoring Report

Severity · WTP / wealth

20%med

share of affected person’s wealth they would pay for a solution

Current solutions

3.0/ 10med

quality of existing solutions — low score = high opportunity

Market size · TAM

$6.0Thigh

USD / year the world is already paying

Time · OOM to impact

12ylow

order-of-magnitude horizon to civilizational-scale impact

Capital · OOM to solve

$200.0Blow

cumulative R&D + deployment + supply chain across the arc

Priority score

16

importance × urgency, 0–100

Importance

18

humans affected × severity, gated by market

Urgency

88

direction of travel + solution gap

Neglectedness

5/10

Education spend is huge, but the high-leverage wedge (scalable 1:1-quality tutoring) is barely funded relative to its potential.

med

Tractability

7/10

Bloom proved 1:1 tutoring lifts outcomes two standard deviations; LLM tutors can now approximate it at near-zero marginal cost.

med

Ways to help

Build

Build an AI tutor with a real pedagogical model and measured learning gains.

Career

Teaching, instructional design, or edtech research.

Organizations

People to follow

Three-lens scoring

welfare · copenhagen BCR
12.0× per $med
x-risk · 80k hours ITNn/a
utility delta · state-of-art vs physics
30%med

Benjamin Bloom's 1984 research showed one-on-one tutoring raises student performance by 2 standard deviations versus classroom instruction. The problem was cost: one tutor per child is infeasible at population scale. AI tutoring changes the economics. 1.5B children are undereducated globally (UNESCO); billions more adults lack skills they could have acquired with better pedagogy. This is the problem Deutsch writes about most directly in The Beginning of Infinity, better explanations compound across generations.

The success vision · 10 years horizon

If we solve this, here is the world we get.

med

Before · today

1:1 human tutoring produces +2σ outcomes vs classroom (Bloom 1984) but is unaffordable for most. Median schooling under-delivers; global learning poverty ~70% in low-income countries.

After · 10 years

Every learner has access to AI tutoring matching the best human tutors, in any language, at near-zero cost. Bloom 2-sigma becomes the default, not the exception.

Voices on this quest

4 thinkers
David DeutschPhysicist & Philosopher · Oxford

Bad explanations waste billions of childhood hours. Good explanations compound across generations. Pedagogy is an epistemology problem.

The Beginning of Infinity, chapter on bad philosophy

Patrick CollisonCo-founder & CEO, Stripe · Stripe

Asks whether we can dramatically improve the quality and speed of learning. Cites tutoring and personalized instruction as high-leverage unsolved problems.

patrickcollison.com/questions

Jason CrawfordFounder · Roots of Progress

Education quality is the compounding lever. Progress requires transmitting knowledge at higher fidelity and lower cost to each successive generation.

Roots of Progress essays

Tyler CowenEconomist & Writer · George Mason, Emergent Ventures

Finding and developing talent is an underrated problem. Most selection is bad, most mentorship is worse, and the tails of ability are where civilizational value compounds.

Talent (with Daniel Gross, 2022)

Companies on this quest

5 mapped
Duolingopublic

Language learning reaching 100M+ monthly users. Expanded into math, music, and literacy.

$14.5Bhigh
Replitprivate

Browser-based coding platform. 100M+ learners worldwide; AI agent for app-building compresses learn-to-ship cycle.

$3.0Bmed
Speakprivate

AI conversational tutor for language learning, unlimited speaking practice at marginal cost.

$1.0Bmed
Khan Academynonprofit

Free world-class education. Khanmigo AI tutor as the flagship attempt at scaling one-on-one pedagogy.

private · no disclosed cap
Synthesisprivate

AI tutor for kids (math, reading). Attempting to close the 2-sigma gap for ages 7-14.

private · no disclosed cap

Capital funding this quest

3 allocators

Writing about this right now

full feed →

No coverage logged yet. The next ingest cron will populate this.

Sources