AI Cracks Centuries-Old Ciphers in Hours, Transforming Historical Research and Code-Breaking

AI Achieves Breakthrough in Historical Cryptography

AI has cracked a 200-year-old cipher ordered by Napoleon Bonaparte in just six hours, a task that could have taken human code-breakers weeks or remained unsolved indefinitely1

1

. Carter Church, a 27-year-old AI engineer at US cybersecurity firm SentinelOne, used OpenAI's GPT-6 Astra to decrypt the 19th-century letter addressed to one of Napoleon's generals2

2

. The message from 1809 detailed French and allied troops' positions across two dozen lines of symbols, numbers, and letters, and had been sitting ignored in an online academic archive3

3

.

Church needed to learn "very little" about code-breaking techniques or the historical period to achieve this breakthrough1

1

. He spent some evenings reviewing the cipher before consulting Satoshi Tomokiyo, a Japanese patent engineer who maintains Cryptiana, an online list of unsolved ciphers2

2

. "What AI mostly brings is persistence," Church told AFP, describing "an unbelievable sense of satisfaction and excitement" upon seeing the result1

1

.

A Turning Point for Code-Breaking Communities

Tomokiyo declared that "2026 will be remembered as a turning point for codebreaking of historical ciphers," noting that since mid-September, most ciphers on his list have been cracked using AI3

3

. The public release of state-of-the-art generalist tools has unleashed dozens of new solutions, with researchers using not only OpenAI's most powerful model but also top offerings from rival US lab Anthropic1

1

.

However, the 58-year-old Tomokiyo expressed mixed feelings about this rapid progress. "I'm excited long-time problems are now solved, but on the other hand, I have mixed feelings, because AI seems to have solved these difficult problems so easily," he said2

2

. One fellow code-breaker told him bluntly: "I feel useless"3

3

. Traditional, low-tech methods remain effective but are time- and labor-intensive, leading historians to leave many ciphers unattended while focusing on more pressing problems1

1

.

AI Transforms Access to Ancient Manuscripts and Medieval Archives

Beyond decryption of historical documents, AI in historical research is revolutionizing how scholars access previously inaccessible sources. The Herculaneum Papyri—hundreds of papyrus scrolls carbonized in the eruption of Vesuvius in 79 AD—are too fragile to unroll without risking destruction2

2

. Researchers in Italy announced in June the first complete virtual unrolling of a scroll using X-ray analysis and AI, revealing a treatise on ethics1

1

.

AI is also transcribing ancient texts at unprecedented scale. A project is building a searchable collection of Jewish history and culture by transcribing 10 million pages of Hebrew manuscripts3

3

. "It's gigantic. No human being could process such a quantity" of data, said project leader Daniel Stoekl Ben Ezra of the Ecole Pratique des Hautes Etudes in Paris1

1

. He likens the databases to a "huge Facebook of the past" that could reveal links between people, places, and ideas2

2

.

Digitizing Medieval Archives Reveals Hidden Historical Insights

At Notre Dame Cathedral archives, an AI system has been trained to transcribe 14,600 Latin pages from medieval archives—a task not dissimilar to code-breaking1

1

. Historian Julie Claustre, who leads the e-NDP project, explains that these are minutes drawn up by notaries with highly administrative, pragmatic script and technical vocabulary3

3

. With character recognition rates between 88 and 94 percent, the project has revealed information about the cathedral's daily life, building work, staff, and care of abandoned children1

1

.

Doing the same work by hand would take "several decades," Claustre noted, but "it's not just a time-saver." Automation enables "inquiries that were not previously possible; it transforms the conditions of research"3

3

. Watch for AI's continued expansion into analyzing millions of historical manuscripts, though Tomokiyo warns that some tough problems could remain enigmas for years because less relevant data is available to help AI figure them out2

2

. As AI surpasses human abilities in pursuits from mathematics to computer programming, the technology is fundamentally altering how we unlock secrets from the past while raising questions about the changing role of human expertise in historical cryptography1

1

.