ideasIdea
Bounties for documented failed replications of high-impact findings
Pay a reward to any lab that carefully tries to repeat an important cancer finding and documents that it did not work. Today that work is unpaid and unpublished.
Security research uses bug bounties to reward finding flaws. A replication bounty programme would list high-impact preclinical oncology findings (selected by citations and translational stage), pay a fixed sum for a pre-registered, adequately powered replication attempt regardless of outcome, and a bonus for the first documented non-replication that passes methodological review. Results are published in a registry.
Hypothesis
The programme will produce at least 50 pre-registered replications per year at a cost per replication under a tenth of a standard grant, and at least a third of targeted findings will fail to replicate.
Rationale
Replication attempts already happen informally in labs and industry; the bounty converts private knowledge into public record at low cost.
What would test it
Fund 50 bounties in year one; measure uptake, methodological quality of submissions and publication of results.
Maturity
speculative
Who has to act
philanthropy
Cost to try
Small (under $1M)
Years to first evidence
1
Bottlenecks it attacks
- Preclinical results do not reproduce · Fewer than half of landmark cancer biology findings reproduce when someone else tries.
- Failures are hidden · Negative trials, failed drugs and abandoned programmes are rarely published, so the same mistakes are repeated.