Development

Rank by exploitation and one model leads; rank by economic value and it reverses

development2026-05-16

Claim

A new exploitation benchmark found Mythos Preview leading at 69% in climbing from vulnerable code to arbitrary code execution, well ahead of GPT-5.5, while ranking the same models by economic value reverses the order with GPT-5.5 winning roughly 98% against last year's champion.

Reported metric: 69% / 98%

Reporting date: 2026-05-16 (not necessarily the event date)

Sources

Attribution and reporting linked by this record; links do not imply independent verification.

Welcome to May 16, 2026 (theinnermostloop.substack.com)

Knowledge graphopen in browser →

Reported in

Welcome to May 16, 2026

Actor

Evidence for

Supersedes

A model writes a console emulator from scratch half the time

Referenced by

Cite or correct this record

Citation formatCitation text

Rank by exploitation and one model leads; rank by economic value and it reverses
Innermost Loop KB — secondary record
Reporting date: 2026-05-16
KB permalink: https://nicholsn.github.io/innermost-loop-kb/developments/2026-05-16-the-rankings-invert-by-what-you-measure
Claim: A new exploitation benchmark found Mythos Preview leading at 69% in climbing from vulnerable code to arbitrary code execution, well ahead of GPT-5.5, while ranking the same models by economic value reverses the order with GPT-5.5 winning roughly 98% against last year's champion.
Attributed sources:
- Welcome to May 16, 2026 — https://theinnermostloop.substack.com/p/welcome-to-may-16-2026
This citation does not imply independent verification of linked-source claims.

Suggest a correction — opens a GitHub draft for you to review and submit.