NOTE: I'm working from memory of some things I read about three years ago.
If I have any errors or omissions please post a correction.

There are two major problems with the Cirrus BRS system at this point:
training and human nature.

In many areas of life, including flight training, we are taught two major
aspects of problem resolution. The first is to identify the problem by using
one or more checklists (written or mental) to determine the precise nature
of the problem. The second is to determine the corrective action, again by
using one or more mental or written checklists. Obviously, if you lose a
wing in flight problem identification is quick, and problem resolution is
limited to one option (prayer). But by and large we use the step-by-step
problem solving techniques we have been taught. That's the training issue.

On the human nature side, we are (by and large) thinking creatures, not just
robots. And our decision making process is often influenced by factors
outside of our training. If you are flying a GA aircraft that you personally
own, in the back of your mind will be the $100,000 or way up you have sunk
in the aircraft. And all too often, in the event of a serious problem, the
end result, if the problem is not corrected, is a broken airplane on the
ground with you in it. So, when your airplane has a problem, these kinds of
factors will be in at least the back of your mind. Any time we are presented
with a problem, human nature drives us to keep trying to find a solution.
And when flying an aircraft, you are even more driven by the unpleasantness
of the possible consequences of not solving the problem.

We have see up the background, now let's look at how this applies to the
Cirrus BRS.

The key factors are the short threshold time, and the narrrow window. In the
Cirrus, when you are presented with certain types of problems, you only have
a short period of time before the chute must be deployed, and a short window
after that when the chute will still be effective. If you delay deployment,
the BRS will not be able to save the aircraft. Let me give you an analogy,
which may or may not be fully accurate, but which will illustrate the point:

Consider a pilot in a jet fighter on an aircraft carrier. He gives his
"thumbs up", the catapult fires, and he begins his takeoff roll. But just
after he rotates, all engines flame out. The pilot then has two options: a
restart, or an eject. If he ejects, he saves himself, but he loses the
airplane. If he restarts, he and the plane both come out OK. But there's a
problem with the restart option: if it is not begun immediately or if it
takes too long, an eject will not work; the chute will not open.

So the pilot is taught that under this set of circumstances he shouldn't
even consider a restart, he should just eject.

And from what I understand, this is the situation with the Cirrus. If you
are presented with a certain type of problem, you must deploy the BRS right
then. You cannot attempt to solve the problem, because if you do, during the
time you spent trying to solve the problem, you have put the BRS outside of
it's operating window and it will no longer function. And you wouldn't be
able to solve the problem in the first place.

So, a Cirrus puts the pilot outside of the problem-solving methods he has
previously learned. The engine stops. In a typical GA plane you check the
fuel, check the mags, check a few other things and try to the resolve the
issue. In a Cirrus, you only need to determine that the engine has stopped
and deploy the BRS. You don't need to know why the engine stopped or what
must be done to restart it: it's stopped, you pull the handle. Obviously,
the "engine out" example is an exaggeration, but you get my point.

Then to human nature. You're in a gypical GA plane, ou're engine is out, you
are going to attempt everything possible to restart it. Because you are
thinking: "I spent $300,000 on this airplane, I'm not going to let it get
bent". And you also have no other alternative. So, you try this, and you try
than, and you try the other until you either get it fixed or you run out of
sky.

But in a Cirrus, you cannot follow human nature, you have to just say it's
broke, pull the handle. In many instances you will not be able to fix what's
wrong, and if you do spend time trying to fix it you will run outside of the
window where the BRS system will properly and effectively deploy.

So, the Cirrus and the BRS system are not inherently less safe than a
conventional aircraft, but you do have to break some old habits and develop
a good understaning of how the aircraft works. "That's broke, pull the
handle", "that's broke, pull the handle", that has to become your mantra.
Then you'll be OK...