I've been tracking the developments since the publication of the article 'Non-Answer on Armed Robot Pullout From Iraq Reveals Fragile Bot Industry' in Popular Mechanics. The article covered the recent RoboBusiness conference and had highlighted the fact that three SWORDS armed robots which had been deployed to Iraq in 2007 and had subsequently been withdrawn. Popular Mechanics cited the US Army's Program Executive Officer, Kevin Fahey, who reportedly said that there was an incident where "the gun started moving when it was not intended to move".
The article didn't go into any further details about the circumstances in question, but that wasn't enough to prevent some hysterical headlines in a number of blogs, for example: 'Combat Robot Attempts Rebellion Against Human Masters in Iraq, Army Pulls Plug for 10-20 Years' (Gizmodo), and 'US war robots in Iraq 'turned guns' on fleshy comrades' (The Register).
Neither article made any reference to the fact that although SWORDS is armed it is NOT autonomous, but is remotely operated by a soldier. The distinction is an important one, as the prospect of armed autonomous robots raises a whole set of issues (as I mentioned in an earlier post 'Autonomous military robots'); whereas, having a 'man-in-the-loop' places the responsibility for making decisions and accountability on the human operator rather than on the machine.
However, it seems that the manufacturers Foster-Miller are concerned about the adverse publicity and have even updated the product description on their website with an editorial comment. (If you want to learn more about the intended role of SWORDS, there's some information on page 17 of The Guardian, Winter 2007 (PDF) and also on DefenseLink).
Interestingly, Popular Mechanics has subsequently published 'The Inside Story of the SWORDS Armed Robot "Pullout" in Iraq: Update' which clarifies the statements made in the earlier article; however, I couldn't help thinking that this presents the story in a slightly different light?