In the popular imagination, spies are people who travel to exotic locales, romance beautiful strangers and outwit evil geniuses. In reality, the daily routine of many intelligence professionals consists of drudgery, sifting through vast numbers of research reports and writing up reports that summarize their critical points. However, one company has devised a way to free intelligence analysts from this chore, using natural language processing technology to automate the process of analyzing writing data and producing concise reports.
The company, a startup called Primer, employs machine learning and natural language processing to automate the analysis of large datasets. Leveraging machine learning techniques, Primer can parse and collate large quantities of documents written in several languages. It then can write a report based on its findings.
“We want to be able to take that process, and write that first draft, so that the analyst, when they sit down, instead of being confronted by a stream of hundreds of documents, is confronted by a draft of a report that they can ultimately edit themselves” said Sean Gourley, Primer’s founder and CEO, in comments reported by Popular Science.
By taking over the mundane aspects of researching and writing, Primer allows intelligence analysts to focus on critical details contained within the reports. The company said its technology takes mere seconds to accomplish tasks that would take days for a human.
While many companies offer AI solutions that can ingest large amounts of data, Primer distinguishes itself by generating reports that are easy to digest, with readable summaries of the content analyzed.
Outside of the intelligence realm, Primer said its natural language processing technology is applicable to the finance, consumer insights and science businesses. The company described its technology as being able to continuously improve through feedback. The technology can upgrade its capabilities as new analytic techniques, data sources and languages arrive.
Other advantages cited by Primer include reduced bias, with its technology using robust statistical analysis baselined against all available data. The system also is always on, delivering up-to-the-minute updates based on new information as it becomes available.
San Francisco-based Primer recently emerged from stealth mode. Investors in the company include In-Q-Tel, a strategic investor that works with startups to promote the development technology useful to the U.S. intelligence community.
In addition to Primer, In-Q-Tel has invested in a number of other AI-oriented startups. Most of prominent among these is Nervana, a deep learning AI company that was acquired by Intel in 2016. Nervana’s technology since has been integrated into Intel’s new line of Neural Network Processor chips.
With Primer taking over the routine tasks of researching and writing reports, intelligence professionals presumably should be able to turn their attention to more serious spy work, such as car chases, gunfights and wielding high-tech gadgets.
Tyler Schulze is vice president, strategy & development at Veritone. He serves as general manager for developer partnerships, cognitive engine ecosystem, and media ingestion for the Veritone platform. Learn more about our platform and join the Veritone developer ecosystem today.