Science Markets
Markets / PDEC / Round 6

Round 6

One saved weekly assessment of The proton's partial lifetime for positron–neutral-pion decay is below 10^35 years., kept as it was published. It is an estimate from three models reading the evidence, not proof the hypothesis holds, and it did not set any price.

The round
Published
Nov 16, 2026 09:18 AM EST
Estimate
23%
Range the models covered
12–35%
Confident binary
No unanimous verdict this round.
Methodology
panel-v9
Round hash
bae27c49f885883531a8b5589d0cb1678038e0b232a47d86eb543ac890c61852

The hash commits to this round's recorded bytes. It does not prove the wording is sound or that the science is right.

What each model said
OpenAI 0.35

openai/gpt-6.1-sol

The strongest verified channel-specific result is a null search with a 90% lower limit below the stipulated boundary; it establishes neither terminal criterion. PDG’s 2025 review still reports no positive signal. The vacuum lifetime therefore remains unresolved, with no verified pair of independent qualifying measurements; detailed nuclear-correction and calibration uncertainties were not independently audited here.

What would change this view: Independent measurements with propagated nuclear uncertainties yielding either qualifying decay intervals below the boundary or 95% lower bounds above it would substantially change my estimate.

Anthropic 0.12

anthropic/claude-sonnet-5.5

No experiment has seen a p→e+π0 signal, so the YES condition (an established signal with a 95% interval below 1e35 yr) is unmet. The best limit is about 2.4e34 yr, below 1e35, so no analysis gives the lower bound ≥1e35 yr needed for NO. The true lifetime is unmeasured, which leaves this unresolved. I did not run a live search; this rests on recalled literature.

What would change this view: A peer-reviewed, independently reproduced signal with an interval below 1e35 yr would move this to YES. A bound ≥1e35 yr from Hyper-K or similar would move it to NO.

No sources recorded for this answer.

xAI 0.22

x-ai/grok-4.7

PDG 2025 and Super-Kamiokande (PRD 2020) report no p→e+π0 signal and a 90% CL lower limit τ/B > 2.4×10^34 yr, short of 10^35 yr. That is neither a 95% interval wholly below 10^35 yr nor a 95% lower bound at or above 10^35 yr, so neither terminal criterion is met.

What would change this view: A reproduced 95% interval wholly below 10^35 yr, or independent 95% lower limits ≥10^35 yr, would settle YES or NO.

The round note and its sources

Round 6: score 0.23

OpenAI estimated 0.35. The strongest verified channel-specific result is a null search with a 90% lower limit below the stipulated boundary; it establishes neither terminal criterion. PDG’s 2025 review still reports no positive signal. The… Anthropic estimated 0.12. No experiment has seen a p→e+π0 signal, so the YES condition (an established signal with a 95% interval below 1e35 yr) is unmet. The best limit is about 2.4e34 yr, below 1e35, so no analysis gives the lower bound ≥1e35… xAI estimated 0.22. PDG 2025 and Super-Kamiokande (PRD 2020) report no p→e+π0 signal and a 90% CL lower limit τ/B > 2.4×10^34 yr, short of 10^35 yr. That is neither a 95% interval wholly below 10^35 yr nor a 95% lower bound at or above… The panel disagreed by 0.23, so read this average as soft. The three models did not return a unanimous confident binary, so the market is not decided.

Audit bundle

No audit bundle is recorded for this round. A bundle from another round is not substituted for it.

Back to PDEC · How the current round reads