3d abstract security team

The idea of law is that it is, at its core, impartial. Decisions are expected to be made in a rational, orderly manner, with the correct result happening every time, and justice being served.

But in the real world, mistakes happen. Evidence can get fouled up, irrational decisions abound, and, crucially, cognitive biases can factor into decision-making in a way that even the most meticulous experts and legal prognosticators cannot discern.

That's why many around the country have aimed to do something about it, including the ABA Center for Innovation and e-discovery software company Relativity. Last week, the pair announced the first phase of the DFNDER (Distributed Forensic Expert Network Delegating Review) Project, aimed at connecting forensic experts, attorneys, and academics investigating potential cases of wrongful conviction across the U.S.

Bryan Wilson, head of the DFNDER Project and a fellow at the ABA Center for Innovation, told LTN at Relativity Fest that the main innovation behind the project is a workflow, built off the Relativity platform, that allows everyone working on the case to come together in one place. Harnessing analytics and review, e-discovery practitioners can review documents, forensic experts can redact potentially biasing information, and forensic auditors can take a look at the redacted information all on the same timeline.

From there, Wilson said, review continues “until you know about whether or not there might have been cognitive bias in the case, whether or not forensics were done really well, and you can compare those against each other to have the data start to tell you where you're most likely to overturn some of these wrongful convictions.”

The DFNDER Project will first focus on two particular areas of forensic evidence that routinely produce invalidated or improper evidence: microscopic hair analysis and arson investigations. According to reports published by the FBI, the National Association of Criminal Defense Lawyers (NACDL), and the Innocence Project cited by the DFNDER Project, FBI testimony on microscopic hair analysis contained testimonial errors in more than 90 percent of the cases.

Wilson said that once the project gets a few cases under its belt, it plans to provide templates to standardize these processes, making it easier for newer attorneys to the space to jump in for pro bono work.

“That's where we want to start, but once we have these templates set up, we're going to be able to continue doing more with bite mark analysis, and other types of forensic analysis,” Wilson added. “It will be easier, and we'll have a road map for how to collect that information and distribute it to people.”

Wilson himself is a recent graduate of the University of Missouri-Kansas City (UMKC) School of Law, where he embraced the intersection of law and technology and its potential power. He came up with the idea for the DFNDER Project while working as a volunteer with the Midwest Innocence Project last November.

“Having worked there before, I had seen what the process had looked like, and it involved a lot of manual intake, a lot of scanning and copying of huge case files,” Wilson explained. “It wasn't very efficient, and it didn't do a very good job of getting these cases moved along in a quick manner.”

Following the idea's formation, a former UMKC professor connected Wilson with Relativity. On its end, Relativity is providing database support for the project, as well as training for Wilson and others to be able to use the tool. Janice Hollman, academic program manager at Relativity, told LTN that the company is continually looking to embrace these types of projects that are pitched to them. Relativity counts 83 schools in its academic program, about half of which are law schools.

The question then becomes, how can it be supported from the software side? Hollman said that the biggest questions in getting the DFNDR Project off the ground had to do with an accessible user interface and streamlined workflow, allowing for seamless transitions of data where users can easily pick up a task where others leave off.

“It's not really crowdsourcing, but it's a large, vast swath of people available to do work,” Hollman added. “It's group collaboration on a huge scale that is going to help people, hopefully, find out whether or not they should be where they are.”

This level of technological innovation isn't always something that pro bono projects embrace wholeheartedly, but Wilson is hoping to help shift that paradigm. He noted that sponsors are needed to show that this technology is capable of indeed minimizing these cognitive biases, and helping find justice through proving or disproving innocence through the data.

As he said, “It's about making everything more transparent and helping the justice system function at a higher level.”

3d abstract security team

The idea of law is that it is, at its core, impartial. Decisions are expected to be made in a rational, orderly manner, with the correct result happening every time, and justice being served.

But in the real world, mistakes happen. Evidence can get fouled up, irrational decisions abound, and, crucially, cognitive biases can factor into decision-making in a way that even the most meticulous experts and legal prognosticators cannot discern.

That's why many around the country have aimed to do something about it, including the ABA Center for Innovation and e-discovery software company Relativity. Last week, the pair announced the first phase of the DFNDER (Distributed Forensic Expert Network Delegating Review) Project, aimed at connecting forensic experts, attorneys, and academics investigating potential cases of wrongful conviction across the U.S.

Bryan Wilson, head of the DFNDER Project and a fellow at the ABA Center for Innovation, told LTN at Relativity Fest that the main innovation behind the project is a workflow, built off the Relativity platform, that allows everyone working on the case to come together in one place. Harnessing analytics and review, e-discovery practitioners can review documents, forensic experts can redact potentially biasing information, and forensic auditors can take a look at the redacted information all on the same timeline.

From there, Wilson said, review continues “until you know about whether or not there might have been cognitive bias in the case, whether or not forensics were done really well, and you can compare those against each other to have the data start to tell you where you're most likely to overturn some of these wrongful convictions.”

The DFNDER Project will first focus on two particular areas of forensic evidence that routinely produce invalidated or improper evidence: microscopic hair analysis and arson investigations. According to reports published by the FBI, the National Association of Criminal Defense Lawyers (NACDL), and the Innocence Project cited by the DFNDER Project, FBI testimony on microscopic hair analysis contained testimonial errors in more than 90 percent of the cases.

Wilson said that once the project gets a few cases under its belt, it plans to provide templates to standardize these processes, making it easier for newer attorneys to the space to jump in for pro bono work.

“That's where we want to start, but once we have these templates set up, we're going to be able to continue doing more with bite mark analysis, and other types of forensic analysis,” Wilson added. “It will be easier, and we'll have a road map for how to collect that information and distribute it to people.”

Wilson himself is a recent graduate of the University of Missouri-Kansas City (UMKC) School of Law, where he embraced the intersection of law and technology and its potential power. He came up with the idea for the DFNDER Project while working as a volunteer with the Midwest Innocence Project last November.

“Having worked there before, I had seen what the process had looked like, and it involved a lot of manual intake, a lot of scanning and copying of huge case files,” Wilson explained. “It wasn't very efficient, and it didn't do a very good job of getting these cases moved along in a quick manner.”

Following the idea's formation, a former UMKC professor connected Wilson with Relativity. On its end, Relativity is providing database support for the project, as well as training for Wilson and others to be able to use the tool. Janice Hollman, academic program manager at Relativity, told LTN that the company is continually looking to embrace these types of projects that are pitched to them. Relativity counts 83 schools in its academic program, about half of which are law schools.

The question then becomes, how can it be supported from the software side? Hollman said that the biggest questions in getting the DFNDR Project off the ground had to do with an accessible user interface and streamlined workflow, allowing for seamless transitions of data where users can easily pick up a task where others leave off.

“It's not really crowdsourcing, but it's a large, vast swath of people available to do work,” Hollman added. “It's group collaboration on a huge scale that is going to help people, hopefully, find out whether or not they should be where they are.”

This level of technological innovation isn't always something that pro bono projects embrace wholeheartedly, but Wilson is hoping to help shift that paradigm. He noted that sponsors are needed to show that this technology is capable of indeed minimizing these cognitive biases, and helping find justice through proving or disproving innocence through the data.

As he said, “It's about making everything more transparent and helping the justice system function at a higher level.”