GoKawiilHistorian tests GPT-6 Sol and Opus 5.5 on unsolved archival puzzles
A researcher describes early experiments using the newly released GPT-6 Sol and Opus 5.5 models to go beyond routine transcription work and attempt to actually solve open historical problems, including tracing alchemical knowledge and decoding 17th century letters. The author outlines criteria for which historical puzzles are 'tractable' for AI, such as codebreaking tasks, citing an example of an AI model decrypting a 1941 German army communication.
GoKawiil's interpretation of the reporting above, not reported fact.
The account suggests frontier AI models have improved enough since 2024-25 to move from assisting historians with document transcription toward proposing or testing actual solutions to unresolved historical questions. The author argues this could prompt AI labs, historians and funders to pursue structured collaborations, though this remains a personal recommendation rather than an established initiative. The framing implies such gains are currently limited to narrow problem types—like cryptography—where answers can be clearly verified.
- New models GPT-6 Sol and Opus 5.5 were tested on historical research problems beyond basic transcription.
- The author identifies criteria—digitized data, verifiability, multilingual reasoning—that make historical problems suitable for AI assistance.
- The piece calls for closer collaboration between AI labs, historians and funding bodies to pursue such research.
Amazon Kindle Scribe — For researchers digging through archival letters and manuscripts, having a device that lets you read, annotate, and organize digitized documents on the go is invaluable. The Kindle Scribe pairs well with the kind of digital humanities workflow described here, letting you take notes alongside AI-assisted transcriptions and analysis.
See Amazon Kindle Scribe on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: resobscura.substack.com — Benjamin Breen, 2026-09-24
Published there as: “Using LLMs to trace alchemical knowledge and decode 17th century letters”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.