"DARPA often looks at the world from the point of view of our potential adversaries to predict what they might do with available technology," said Improv's project manager, John Main, in a statement.
Improvised explosive devices and other homemade weapons, which were first used in the 1500s, have traditionally been considered strategically ineffective, yet modern ones, made with everything from washing machine parts to transistor radios to video game controllers, killed or wounded more US troops in Iraq and Afghanistan than any other weapon.
In certain countries, "there are a lot of people with engineering backgrounds, and not a lot to do," says Ben FitzGerald, director of the Technology and National Security Program at the Center for a New American Security (CNAS).
However, "the next IED won't be an IED," FitzGerald said. "It will be some other technology we didn't identify, and that's something we have to deal with."
Improv's overall emphasis will be on speed and economy, and the agency is particularly interested in projects that use "transportation, construction, maritime, and communications" technologies for projects, the project application says.
"No proposals involving human or animal research will be considered," but participants are "encouraged to utilize open source code," the application says.
Of the submitted proposals, DARPA says it will fund a set number of projects for short feasibility studies. Applicants will get up to $40,000 to refine their ideas. The most promising concepts will then be given 75 days, and up to $70,000, to make a prototype. Winning prototypes will receive $20,000 for evaluation at a US government test facility.
The ideas that come out of Improv may do more than give the Department of Defense a sense for what kinds of as-yet unknown weapons it will be up against. Improv could also reveal "new concepts of operations using existing, off-the-shelf technology," says Thomas Dietterich, a pioneering researcher in artificial intelligence who has worked on a number of past DARPA projects.
It's something the US Air Force did in 2010, when it built the fastest supercomputer in the entire DoD — 33rd fastest in the world at the time — out of 1,760 Playstation 3 consoles. Along similar lines, Dietterich, who worked on a DARPA-funded program in the mid-2000s that became the basis for, among other things, Apple's Siri personal assistant, can envision a fully-armed "semi-autonomous weapons system that is small and fast for use in urban environments" materializing from little more than some consumer-grade drones, a 3D printer, and some artificial intelligence.
"For a long time, DoD has pursued a top-down process of acquiring weapons systems in which a design team writes the specs and then they are put out for bid," Dietterich says. "This DARPA program is flipping that around and asking the maker community to come up with weird and creative ideas."
Ideas have been generated for the military by civilians before. As John Chambers points out in the Oxford Companion to American Military History, almost all of the most important military devices and weapons over the past two centuries — including the automatic machine gun, the helicopter, and the modern submarine — originated with civilians working without any mandate or request from the military.
The conventional industrial model governing the American defense industry is severely hamstrung by everything from the vagaries of funding cycles to the influence of corporate lobbyists, making it tremendously difficult for the US military to keep pace with its nimbler adversaries.
With little incentive to change, old-line defense contractors compound the problem by continuing to "think in traditional ways," says John Clippinger, a research scientist at the MIT Media Lab Human Dynamics Group.
Clippinger, who has helped develop network-centric warfare concepts for the DOD over the past decade-and-a-half, believes that "innovation networks" like the one DARPA is trying to build with Improv are the best, perhaps even only, way to keep pace with a loosely-coupled, quickly-organizing terrorist group like ISIS.
However, there are segments of the population with which DARPA wants to work with that Clippinger doesn't think will necessarily want to work with the agency.
"I have done research through DARPA, and there's some very creative thinking there," he says. "But when you get pulled into the classified world, things suddenly go to a dark territory. Which is one of the problems with DARPA for academics and certain other people, frankly."
And, while DARPA may get submissions from plenty of scientists, engineers, and "skilled hobbyists," will top-tier criminals and terrorists will be eager to share their best ideas with the US government?
If they are, the question remains: will DARPA want to work them? Dzmitry Naskovets, a convicted cybercriminal sentenced to 33 months in federal prison in 2011 for conspiracy to commit wire fraud and credit card fraud, reached out to DARPA while in Brooklyn's Metropolitan Detention Center.
DARPA was created by President Dwight Eisenhower in 1958 in response to the Soviet Union's launch of its Sputnik satellite the year before. It has received occasional criticism over the years for some of its weirder initiatives, including experiments that investigated the use of neuroscience to control people's minds, and, like many federal agencies, DARPA is not immune to the odd ethics violation. On the other hand, recent breakthroughs by DARPA researchers have included handheld paddles that enable soldiers to climb vertically like geckos, bullets that can be maneuvered while in flight, and prosthetic limbs that function almost exactly like real ones.
In practical terms, the idea of freelance bomb makers vying to win a government-sponsored competition would seem to raise a number of legal concerns.
"How do you at once stop bad things from happening, yet encourage them at the same time?" wondered Clippinger. (Improv project manager Main did not respond to requests for comment, and DARPA declined to make other agency officials available for interviews).
DARPA's rules say simply: "Proposers are free to reconfigure, repurpose, program, reprogram, modify, combine, or recombine commercially available technology in any way within the bounds of local, state, and federal laws and regulations."
It seems awfully tough for amateur inventors or hobbyists to develop would-be threats to the military without at least inadvertently committing a number of serious crimes along the way — which also poses serious problems for emergency responders, says John Jay College of Criminal Justice Professor and ex-NYPD Detective Sergeant Joe Giacalone.
The idea that DARPA would ask regular citizens to create dangerous devices also strikes personal injury lawyer Gabriel Levin as "very odd." Aside from the obvious physical dangers, a program like Improv would appear to open the government up to all manner of liability issues — and as Levin says, "injury" can also mean financial damage done by a computer virus that accidentally gets out.Whatever happens, Larry Johnson, a former CIA officer who served as deputy director of the State Department's Office of Counterterrorism from 1989 to 1993, isn't buying any of it. He says he sees Improv as "white man's welfare," which may generate some interesting ideas, but will ultimately end up benefiting the same big corporations that were awarded $272.4 billion in defense contracts last year.Says Johnson, "It will be declared a great victory, and everybody will say what a great public-private initiative it was, but all I see is another back door to create another business opportunity for defense contractors."