Machine Learning

Investigators are using AI to solve the mystery of Anne Frank’s betrayal

©iSTOCK/LordRunar

A team of forensic investigators are using advancements in AI to hopefully solve the cold case of who betrayed Anne Frank back in 1944.

Everyone should know the harrowing story of Anne Frank, but one missing piece never uncovered is of who betrayed the young girl which led her to death in a concentration camp. Historians, journalists, and investigators have spent decades attempting to uncover the betrayer’s identity with little success.

Anne’s father, Otto, was the sole survivor of the group of eight people who hid in the annex of his company before their capture. When Otto returned from Auschwitz, he called for an investigation into who betrayed his family.

A strong suspicion was towards warehouse worker Wilhelm van Maaren; although an investigation by the Dutch police in 1947 uncovered no conclusive evidence against him. A second investigation conducted by the police in 1963 also failed to find Wilhelm guilty.

In the decades since, around 30 suspects have been accused. These range from neighbours who may have seen the people in hiding, a former employee who suspected Otto of having an affair with his wife, to Jewish woman Ans van Dijk who became an agent for the Gestapo and was linked to dozens of betrayals.

In 2003, the Dutch Institute for War, Holocaust and Genocide Studies concluded there was not enough conclusive evidence to implicate any of the four leading suspects. Yet another theory was put forward just last year which theorised no betrayal took place and a happenstance raid discovered the people hiding, but further investigation was recommended. Basically, no-one knows.

Using AI to make a breakthrough

To break this impasse in the investigation, a special algorithm has been developed to scour hordes of documents. The AI is being fed with known Nazi collaborators, informants, historic documents, police records, and prior research.

“Our software allows [the team] to search the data and visualize it new ways. This has already led to a few new traces,” said Marius Helf, chief data scientist at Xomnia. “In the future, we plan to make the systems more intelligent, in the sense that it will be able to automatically connect persons, events, and places.”

Retired investigator Vincent Pankoke told Reuters the software could provide new leads and connections based on the trove of data “that a human in their lifetime might not be able to review.”

While even the best AIs are currently beat by six-year-old humans, computers have the ability to process data much faster.

Pankoke is hoping by working with Xomnia they will be able to finally close the case by August 4th, 2019 which marks 75 years since the arrest of Anne Frank.

The effort is being crowdfunded and the project is looking to raise $5 million. If you can help out, you can find donation information here.

Do you believe AI is able to make breakthroughs in cold cases? Let us know in the comments.

Click to comment

You must be logged in to post a comment Login

Leave a Reply

To Top