While I don't have a high opinion of the SLS stack or of Congressional pork... One of the problems NASA is faced with is justifying their existence, which in the public eye (i.e. our representatives in Congress) has to include some form of manned spaceflight. This while not having the funding wherewithal (as provided by the same Congressional representatives) to complete a flagship manned planetary or lunar exploration program.
You can't completely discount manned spaceflight as having absolutely no scientific value; because, there is scientific and technical value in performing a manned mission to an asteroid parked in orbit. There's a quote that goes something like: "A geologist with a rock hammer could learn more in an hour than the robotic probe learned in a year." It's a gross oversimplification but there's some validity to it. By putting people on the site, we can learn quite a bit about the makeup and structure of the sample asteroid that might not be possible with a single pre-designed robotic probe, regardless of how well designed. A human in-situ can make observations and connections based on experience that a robot might not be equipped to make. Said human could then make decisions based on that information, which might be different than decisions made by a remote operator.
In addition, in the process of designing and flying the vehicle we gain valuable technical information necessary to support various human. We get improved engine designs for vehicles, human or robotic. We get better life-support designs which can be applicable to terrestrial applications as well as space. We get better harsh environment suits. We test the ability of our systems to support "long-endurance" human spaceflight outside the cozy protected orbit ISS resides in.
Now, if your argument is simply that we shouldn't be putting humans into space at all, as it's too dangerous and/or expensive... I don't know that I have a counter argument that you'll find acceptable. Sure, we can do some things more cheaply using robots and automation. That's true with many human endeavors. McDonalds is working on automating food production in their restaurants, for example. Agriculture can be largely automated. Manufacturing can be automated. Art can be automated. We use human labor because humans are cheap(-er) than using machines or humans introduce artistic / cultural variation, or humans can make observations and connections based on experience that machines cannot (yet) make, or we just have a surplus of humans looking for something to do...
Sending people to space is expensive because of the energy cost (boost out of the gravity well) and life-support cost. There's no getting around the energy required to boost out of the gravity well; but, we can improve the vehicle and drive costs down. Life support costs we can reduce through infrastructure improvements. If we're ever to have any kind of presence in space, we have to start somewhere; and, that's going to require putting humans into space to obtain experience that cannot be obtained remotely via robotic telepresence. If we're going to reduce the cost, we have to build infrastructure, which is largely going to require some human supervision. We're going to go there eventually; the root technology is available now.