AGIdex
agidex/Capabilities/Creativity/Novel Idea Generation
SUB-CAPABILITY · CREATIVITY

Novel Idea Generation

46%
vs. HUMAN BASELINE = 100
PartialConf · Low

Scoring rubric

0–20
Early research
20–40
Narrow benchmark competence
40–60
Strong benchmark · reliability gaps
60–80
Human-competitive in common scenarios
80–100
Comparable to typical skilled adult
100+
Reliably exceeds typical human baseline

Score vs. baseline

Trend · Last 12 months
12mo ago   %
6mo ago   %
Today   %
Δ 12mo   +0

What this measures

Human baseline

Combine existing ideas into proposals not obviously templated or derivative.

Human frontier

Produce ideas a domain expert would recognize as genuinely new contributions.

Current state

What works

Brainstorm-style ideation at scale; surface variety across many domains.

Key gaps

Distinguishing recombination from genuine novelty remains unsolved even by evaluators.

Evidence

0 sources
TechnologyQualitySourceScore vs. baselineScore
No evidence records yet.
Source: paperswithcode.com/task/open-ended-generation · Last reviewed May 2026