Must have:

• A marginal cost to a military/non-state-actor of less than $1000 (2025 dollars)

• >=20km range

• >=1kg weaponized payload

• Ability to function properly without any radio signal or other external communication, excluding short-range radio-based friend-or-foe identification or communication with other drones

• Ability to identify targets and navigate towards them - i.e. no artillery, but certain cruise missiles count. Target identification may be arbitrarily simple, based on heat signatures or simple visual identification.

• Expectation of achieving an unambiguously >10% hit rate against either humans or vehicles while operating under the above criteria and in real-world combat environments without active countermeasures (other than radio jamming). Targets may be either pre-selected or selected autonomously on the fly.

In order to settle YES, I must:

• Find a specific drone which matches these criteria

• Have at least one reliable source attesting that the drone can operate fully independently of human control and external communication

• Be highly confident that it matches the other criteria. That is, if we don't have exact sources on cost, range, etc but can easily extrapolate that information from other characteristics of the drone the market can settle YES.

At the beginning of 2028 I will provide a draft settlement, then wait 2 weeks for participants to check my work and volunteer considerations I might have missed.

The drone doesn't need to have ever been actually used in combat, or have entered mass production (though the latter is likely necessary for meeting cost requirements). It also need not be reusable.