Stay up to date with notifications from The Independent

Notifications can be managed in browser preferences.

Robot investigators 'could be used to examine documents in criminal cases'

Decision comes after technology was used to analyse cache of 30 million documents disclosed by Rolls-Royce during major investigation

David Hughes
Thursday 14 December 2017 02:22 GMT
Comments
Artificial intelligence could make some tasks easier
Artificial intelligence could make some tasks easier (Shutterstock)

Robot investigators could be widely used in future to examine documents in complex cases, the head of the Serious Fraud Office (SFO) has suggested.

David Green said he would like to see the possibility of employing artificial intelligence "carefully examined" after using technology to sift through a cache of 30 million documents disclosed by Rolls-Royce during a major investigation.

He also said it was now "pretty clear" that his agency would continue as an independent body after the government dropped plans to have it taken into the National Crime Agency.

The SFO director set out how the Rolls-Royce documents had been examined by a computer algorithm which had the ability to learn as it went along.

The technology was trying to find legally privileged documents which could not be used in the case, but Mr Green suggested that in future similar methods could be used to identify useful evidence in investigations.

He told MPs: "During the Rolls-Royce investigation, which continues because we haven't yet made charging decisions on individuals, Rolls-Royce the company showed what was described in court as extraordinary co-operation.

"A measure of that is that they referred to us a chunk of 30 million documents, mainly digital.

"In doing so they said 'We are well aware that this material does contain material subject to legal professional privilege' which we are not entitled to have and we have to quarantine.

"They agreed with us that we could search the material and sift that out. They also agreed with us that we could use artificial intelligence to do that task."

Using the algorithm "it probably took a tenth of the time it would have taken" using the normal method, which would have involved using around 30 lawyers to sift through the material.

"Also, we found it was more reliable than human intervention," he told the Justice Select Committee.

He suggested that "in time, the criminal courts will be looking towards this methodology", particularly in relation to the disclosure of evidence by the prosecution to the defence.

"The other interesting angle is actually in investigation, with a view to sifting out for relevance and so forth, which would be a not dissimilar thing."

Mr Green told MPs: "We haven't got robots working yet, we've used them once in connection with Rolls-Royce. It's very much a possibility and an aspiration for the future."

Committee chairman Robert Neill asked: "You are certainly convinced that robotics offers real opportunities?"

"Absolutely," Mr Green replied.

In Theresa May's election manifesto, the Tories vowed to incorporate the SFO into the National Crime Agency.

But the recently published economic crime strategy only set out plans to give the NCA power to direct the SFO.

Mr Green, who will stand down in April, said: "I think it's pretty clear that the independence of the SFO will continue."

He said he was "pretty relaxed" about the NCA being given the power to direct the SFO, adding that it already had that relationship with police and had only used it once since 2013.

"In all honesty, quite what this adds, I don't know. But insofar as it is intended to increase and guarantee co-operation between the many agencies on the economic crime waterfront, who could possibly object? I certainly don't," he said.

"I do not think that this power of direction impinges on SFO independence."

PA

Join our commenting forum

Join thought-provoking conversations, follow other Independent readers and see their replies

Comments

Thank you for registering

Please refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged in