AaronP/Bauer-Griffin/GC Images
To say Ancestry’s database is vast would be an understatement.
“We’ve collected over 65 billion records across 80-plus countries,” Sriram Thiagarajan, the company’s chief technology officer and executive vice president of product and technology, told Business Insider. “Just to give a scale, that’s about 10,000 terabytes of data on our platform that we use to provide discoveries to our users.”
The Utah-based genealogy company, founded in 1983, collects records to help people unearth their family roots.
Those records include birth, death, marriage, census, military, land, immigration, and newspapers. Ancestry, which also offers consumer DNA test kits, collaborates with institutions like the National Archives and Records Administration to collect that data.
There’s a daunting caveat with a trove that size, though: organizing it.
Thiagarajan said Ancestry is leveraging AI and machine learning to make the Herculean task easier.
Streamlining with computer vision
When he joined Ancestry’s team in 2017, Thiagarajan said the company had just begun exploring AI and machine learning.
“We were trying to figure out an effective and efficient way to digitize content that we acquire from around the world,” Thiagarajan said.
Ancestry used to scan the records, then outsource operations to manually index and key relevant fields. That information was uploaded into Ancestry’s database before software programs established relationships between people, locations, or other categories.
“About 15 or 20 years ago, when we digitized the 1940 census, it took us about nine months to do it in a manual way at 10 times the cost,” Thiagarajan said.
That sent the Ancestry team searching for answers.
“We said, ‘Why don’t we apply computer vision AI techniques to automatically digitize content without manual intervention?’ Thiagarajan said. “Fast forward to the 2021 timeframe, we used our own proprietary handwriting recognition computer vision technologies, and we compressed the time to market to under nine days from nine months at a fraction of the cost.”
Thiagarajan said Ancestry has since expanded that technology to process other record types, but said humans still review AI results “as needed.”
“We’ve built some automated controls and systems that certainly reduce the amount of time we need to spend checking,” he said. “We want to be extra careful in making sure that what we produce using AI is grounded in truth. Grounded in facts.”
Thiagarajan added that the “extent to which we do it now versus a couple of years ago has certainly improved.”
“At the end of the day, when consumers come to our platform looking for stories about their ancestors, we want to connect them with the records we find,” he said.
Ancestry is beta testing a new AI feature
In addition to implementing AI into the backend, Ancestry has deployed several features for users, including its handwriting recognition tool. It began testing an AI assistant in 2024.
More recently, Thiagarajan told Business Insider that Ancestry began beta testing an AI-powered feature called Audio Stories that allows users to turn records into a narrative audio.
“Our AI can understand the context between the printed material, the image, and the handwritten narrative, and tie it all together into a story,” Thiagarajan said.
Although there’s no official launch date for Audio Stories yet, Ancestry is already thinking above and beyond audio.
“Down the road, we want to be able to add sight, sound, motion, and video storytelling,” Thiagarajan said.
The post Ancestry used to need 9 months to scan census records and make connections for users. AI cut it to under 9 days. appeared first on Business Insider.
