γ
Task coverage
How much of the complete research chain can the machine perform at the required quality?
Endogenous growth · idea production · artificial intelligence
Romer After
the Machine
In his 1990 model, Paul Romer made technological change the result of purposeful research. The machine now enters that idea factory—but it does not make proof, absorption, or diffusion free.
A model can generate a thousand hypotheses before breakfast. Growth still arrives one validated idea at a time.
Research cut-off · 5 August 2026Interactive reconstruction · model outputs are illustrative, not forecasts
Romer 1990 · reconstructed
The loop that moved technology inside economics
Select a stage. Romer separates the reusable design from the rival object built with it, then lets the accumulated stock of designs make tomorrow's research more productive.
Research sector
Purposeful effort
Research human capital combines with the inherited knowledge stock. Existing ideas are both an input into discovery and the record of solved problems.
New knowledge feeds back into the next research cycle
The AI extension
Four parameters decide whether the curve bends
Capability is only one input. Research output is shaped jointly by breadth, relative performance, the O-ring structure of the process, and the rate at which plausible output becomes trusted knowledge.
M
Machine productivity
How large is the cost, speed, or quality advantage on tasks that are technically feasible?
θ
Bottleneck strength
Can exceptional performance on one stage substitute for weak performance elsewhere?
V
Validation and absorption
What share of generated output survives proof, engineering, governance, adoption, and diffusion?
AI research funnel · illustrative
Generation is not yet knowledge
AI can widen the top of the funnel. The growth effect is governed by throughput at the narrowest complementary stage—not by proposal volume.
Shadow price moves to
Experiment and replication
- Stage 01
Generated candidates
Starting candidate pool
1,000
100% of initial
- Stage 02
Screened proposals
58% pass-through from prior stage
580
58% of initial
- Stage 03Binding drop
Experiments / simulations
53.4% pass-through from prior stage
310
31% of initial
- Stage 04
Validated knowledge
58.1% pass-through from prior stage
180
18% of initial
- Stage 05
Adopted innovation
55.6% pass-through from prior stage
100
10% of initial
10%
Final yield
900
Candidates filtered
53.4%
Binding pass-through
Illustrative conversion from 1,000 machine-generated candidates
Live model · not a forecast
Run the idea factory
The same machine can produce a modest level shift or a persistent acceleration. The difference lies in task coverage, complementarity, validation, and how strongly new knowledge raises future research productivity.
1.61×
effective research tasks
138
knowledge index · year 40
0.81%
annualized knowledge growth
46%
candidate survival
Candidate flow
2.31
plausible additions per model period
Validation gap
1.25
candidates that do not enter usable knowledge
Continue into the full argument
The Idea Factory Meets the AI Factory
The essay reconstructs Romer, extends the research production function, reviews the newest evidence, maps the expanding idea space, separates levels from growth rates, and proposes a falsifiable empirical programme for validated knowledge.