One of the most controversial aspects of the Iraq war was the heavy use of defense contractors, who were in many cases paid vast sums of money to do jobs that you’d think an ordinary soldier could do. When it was all said and done, defense contractors had reaped $138 billion dollars by providing security, logistics, and construction services. Among the most notorious of these contractors was Blackwater, whose employees gained a reputation for reckless behavior that caused many unnecessary deaths.

Fast forward to today, and now Blackwater’s founder, former Navy Seal Erik Prince, is pushing for a plan to win the war in Afghanistan by replacing the soldiers with defense contractors. Prince first suggested the plan last May in an op-ed in the Wall Street Journal, where he described this idea in colonial terms. The private military units would be based on units that were deployed by the British East India Company, and would be lead by a single person who he referred to as an “American Viceroy,” that would report directly to the president.

As strange as it may sound, Trump appears to be taking the idea seriously. It’s hard to blame him. Afghanistan is now America’s longest running war, and no matter how many soldiers, generals, aid, or money we throw at the country, nothing seems to bring stability.  Which is why this radical new plan probably looks very enticing to Trump right now.

Under his proposal, private advisers would work directly with Afghanistan combat battalions throughout the country, and the air force would be used for medical evacuation, fire support and ferrying troops.

Prince said the contractors would be “adjuncts” of the Afghan military and would wear that nation’s military uniforms. Pilots would only drop ordnance with Afghan government approval, he said.

Currently, troops from a U.S.-led coalition are stationed primarily at top level headquarters and are not embedded with conventional combat units in the field. Under the plan the contractors would be embedded with Afghanistan’s more than 90 combat battalions throughout the country.

In addition the plan will supposedly cost $10 billion per year, which is a far cry from the $40 billion that is currently being spent. It would involve replacing much of the 8,400 troops who are currently stationed in Afghanistan with 5,500 contractors, and provide a privately owned air force of 90 planes that wouldn’t give fire support without the permission of the Afghan government.

However, there is one problem with this plan. There’s probably a good reason why our soldiers aren’t embedded with the Afghan army in such a decentralized fashion. It probably has to do with the fact that our soldiers are routinely killed by Taliban sympathizers masquerading as Afghan soldiers (many of whom are severely demoralized, poorly paid and equipped, and lead by corrupt officers). Without separating the contractors into heavily fortified compounds (as our military does right now), and by spreading them thin across a vast and desolate country, these contractors would be extremely vulnerable.

It’s certainly a unique plan that is at least different from what our military has been doing in Afghanistan since 2001, which has been more of the same, year after year. But it’s also fraught with danger and the potential for disaster. This is Afghanistan we’re talking about. When it comes to military operations led by advanced nations, disaster is always the safest bet. After all, they don’t call Afghanistan the “graveyard of empires” for nothing.

Perhaps instead of asking how we could win the war in Afghanistan, we should be asking ourselves how we can get out as fast as possible and cut our losses. It’s been 16 years, and there’s no sign that the people in that country have any interest in becoming the kind of liberal democracy that we want them to be. And without that support, there is no military plan in the world that will ever give us a satisfying victory in Afghanistan.