To try and establish a set of rules for UAV airworthiness is akin to writing a set of rules for everything from a paper dart to an Airbus A380.
At one extreme are the indoor ornithopters and hover platforms and at the other, Global Hawk.
There are a lot of misconceptions that UAVs only use the latest technology. One of the more successful ones in current service uses lead acid gelcells for onboard power and mechanical gyros. The flight limitations take account of the hardware used.
Generally a designer has a idea of the flight regime and designs the airframe and selects the avionics accordingly. If QinetiQ's Zephyr 7 had been built to a standard airworthiness and avionics packaging requirements it probably wouldn't have left the ground, let alone stayed aloft for two weeks.
Getting back to the original question, yes it is possible to overstress airframes if the onboard autopilot doesn't have limitations set. A decision made on the ground cannot be relied upon as it will probably be too late.
Watch the recovery of a Scan Eagle and you will appreciate that flight loads are the least of its worries.