Amazon “reviewing” algorithms that promoted bomb materials
Amazon has told the BBC that it is reviewing the way products are presented on its site, after it was found that the service’s algorithms were grouping material that could be used to create explosives.
A Channel 4 News report found yesterday that Amazon’s “frequently bought together” algorithms could be used to help guide would-be terrorists in creating explosives.
When adding widely available chemicals to their shopping basket, investigators for the channel found the site recommended other ingredients for making thermite – used in incendiary bombs. Some of the listings for these chemical components also included recommendations for steel ball bearings, push button switches and battery cables in the algorithmically generated “Customers who bought this item also bought” section.
The BBC asked Amazon how many people would have needed to make the purchases for them to be grouped in this way, but the company refused to give details. Ex-Amazon software engineer, Owen Miller, did however tell the broadcaster that it would have to be a “significant number”.
“All products sold on Amazon must adhere to our selling guidelines and we only sell products that comply with UK laws,” Amazon told the BBC.
“In light of recent events, we are reviewing our website to ensure that all these products are presented in an appropriate manner.
“We also continue to work closely with police and law enforcement agencies when circumstances arise where we can assist their investigations.”
While Amazon has not said how it plans to do this, one option would be for specific recommendations to be filtered.
UK biggest source of clicks for jihadi content in Europe, report finds
Today it was also revealed that online jihadi content attracts more clicks in the UK than any other country in Europe.
Analysts at Policy Exchange found that the UK is the fifth-most-frequent location where jihadi propaganda is accessed, following Turkey, the US, Saudi Arabia and Iraq. Considering the US has a population almost five times that of the UK, that’s a pretty notable finding.
The think tank’s report calls for “responsive regulation” of technology companies, with the government putting political and financial pressure on companies such as Google and Facebook to counter extremist content. This echoes recent rhetoric by Theresa May, who met with France’s Emmanuel Macron in June to discuss a potential collaboration for curbing the spread of jihadi propaganda.
Policy Exchange claims ISIS produces around 100 pieces of new content each week, adding that this is a conservative estimate, and that the new material builds on an archive spanning more than three decades.
“It is clear that our counter-extremism efforts and other initiatives to combat extremism online have, until now, been inadequate,” writes US Army general David H Petraeus, in a foreword to the report. “In fact, I do not think we have yet developed all the ‘big ideas’ needed to come to grips with the problem, much less the policies and methods to combat it.”
He says that last week’s Parson’s Green Tube attack “merely underscored once again the ever-present nature of this threat”.
The report, titled ‘The New Netwar’, puts forward a suggestion for legislation similar to that used in combating child pornography, with a scale of penalties depending on the level of possession and consumption of criminal material. Under section 58 of the Terrorism Act 2000, it is an offence to possess information that could be used to assist a would-be terrorist, but not content that glorified terrorism.
Despite the loss of ISIS territory in the Middle East, the report found that the production of jihadi content has continued and that it has been “largely consistent over the long term”.
At the end of last year, Facebook, Google, Twitter and Microsoft announced a collaborative effort to combat extremist material on their platforms, sharing a database of “hashes” – unique digital footprints – for prohibited content. The findings of Policy Exchange’s polling, however, suggest these measures aren’t doing much to reassure the public. A survey of 2,001 adults in the UK found that 74% felt that the big internet companies should be more proactive in locating and deleting extremist content, and 65% felt the major internet companies are not doing enough to combat online radicalisation.