The AI helped write the scripts that tested a puzzle image for hidden data
A puzzle solver designed tests for whether a circulated image hid steganographic data, and used AI to help write the scripts and qualify signal and noise across JPG compression formats. He ran the tests with standard tools on every copy of the image he could find, and nothing extracted from any of them, which settled that the image itself carries no hidden payload.
How it works
- He designs the test approach himself
- The AI helps write the scripts and the noise and signal qualification across JPG formats
- He runs steghide and outguess on every copy of the image he can find, with several candidate keys
- Nothing extracts from any copy
- He posts the null result with the supporting analysis
Why it’s worth knowing
The useful result was a well supported nothing. The agent built the test rig that let one person check every copy of the image instead of guessing.
Where it’s from
Summarized from a public post on puzzling.stackexchange.com. See the original ↗