There are already a half a dozen or so cars that are full battery powered for under forty thousand, or less than half the price of the current transaction prices of the S. The luxury market is safe territory because the margins are large enough for Tesla to fit within.
As for interior work, I am not sold on a full size touch screen. For the most part it looks a bit tacky to me, its just too "there". That and I am a dial person, specifically for radio and such. I just like the idea of feedback. Sure I could talk to my car and one day that may work when I don't have to drive it but I am not ready to give up that fun.
I don't see how that benefits the incumbents. Unfortunately, the realities of physics and economics (to say nothing about total pollution generated) mean that there are really only two choices: Burn liquid hydrocarbons; or store electricity in a battery (or some combination of the two). Tesla is currently pursing the second option and traditional auto manufacturers are using both. Hydrogen, natural gas and compressed air have all the limitations of batteries and more: The technology that goes in the car is more expensive, the infrastructure is more expensive and exists in fewer places (excluding natural gas), range is short, and 'refuling' times are long.
About ten years ago, as an undergrad, I worked on an NSF- and DOE-funded project to use single-walled carbon nanotubes both as a catalyst in hydrogen fuel cells and for hydrogen storage. Unfortunately, the efficiency gains were minimal, and I'm not aware of much progress that has been made since (but I now work in a different sub-field). That means that a practical fuel cell, big enough to power a lightweight vehicle still costs hundreds of thousands to millions of dollars (mostly due to the requirement for large amounts of platinum to catalyze the H2 in to 2H). Frankly, there has been enough money wasted on pie-in-the-sky projects when there is already-existing technology that works (lithium-ion batteries).
If you don't trust a random stranger on the Internet, then please trust the managers of auto companies. Honda has effectively abandoned their FCX Clairity project because it is cheaper to meet CARB (CA Air Resources Board) requirements by buying zero-emissions credits from Toyota, Tesla, or GM. The FCX, by the way, was a hybrid that only used a fuel cell to meet the average load for electricity and batteries to meet the peak load.
 You need at least an order of magnitude better catalyst performance before you can even consider using PEM fuel cells in a mass-produced car. Every time I've seen a press release about better fuel cells, they seem to mention a few percent. At the current rate of improvement, it will take a quarter of a century before we even solve one of the show-stoppers for hydrogen cars.
Compressed natural gas/biogas is available today and has been for years, at prices far below that of electric cars, with none of the range and refueling issues that electric cars have.
We still import 4 million barrels per day from OPEC. But now, we have the capability to stop all oil imports from OPEC within 60 months. We have low cost natural gas and low cost technology for converting cars to operate on CNG. This program would convert 65 million vehicles (23% of our fleet) to (CNG). Cost $98 billion. The other part of the program is to build 10,000 CNG refueling stations. Cost $20 billion. Total $118 billion. All the costs will be money spent on U.S. labor and material. Use of low cost natural gas will save us about $80 billion per year.
The program can start immediately by presidential order to convert the 600,000 federal non-military vehicles to CNG. Theses are shovel/wrench ready projects. Total cost: less than $5 billion.
This CNG program is not like the Manhattan Project that involved large technical uncertainties and risk. CNG technology is commercially available in the United States. Iran now has 2.9 million vehicles (23% of its fleet) operating on CNG.
The collateral benefits are manifold: cost savings; reduction in trade deficit; employment for 100,000 Americans; reduced CO2 emissions; low technical, commercial and environmental risks; progress that can be accurately measured; plus no political party would find it objectionable.
Eye-opening to say the least.
The second problem can be partially ameliorated by filling at home with a high-pressure compressor (runs on electricity) fed by the same pipe that brings natural gas for your furnace and stove. Unfortunately, that still requires you to plug in the car overnight, just as you would with an electric, because fast high-pressure compressors are still very expensive. The compressor, by the way, costs more than an 80A EVSE for an electric car and adds range more slowly. There are a few public stations that store high pressure gas, but they tend to be at places like airports and taxi maintenance yards, where fleet vehicles congregate (check out Edmund's review of the Civic GX for refill times, though, they aren't nearly as fast as with gasoline). And compressing the gas still requires electricity. It is comparatively easier (and cheaper) to set up an EVSE (the "charger" stand, although it isn't really a charger) for an electric car than to run a gas line and install a gas compressor.
The first problem I mentioned doesn't really have a solution with a short time horizon: You need either more volume (a bigger car) or a tank that can hold more pressure (you think that everyone from SCUBA divers to NASA hasn't been working on that one for the last fifty years?).
Liquefied natural gas has it's own set of problems: people complain about the Tesla Model S losing a few miles of range while parked overnight, but the boil-off from a cryogenic container would be far worse (an uncovered dewar of liquid nitrogen boils off in "a few" hours; a big 150L dewar at 1 bar will last maybe two or three weeks).
It's not that I think natural gas is "bad," it's that as a physicist I can see that we already have technologies that work better (albeit slightly more expensively for now). The only thing natural gas seem to have going for it at the moment is that it is slightly cheaper than a battery-powered car. Unless there is a new and obviously-practicable way to greatly improve the inferior technology in a short amount of time, I don't understand why it is worthwhile to spend money developing it.
 The Civic GX seems be to only available in jurisdictions (like California) that require a manufacturers to sell a certain number of "zero emissions" vehicles. Thus, it is probable that Civic GX's are "compliance cars" and that Honda takes a loss on each one sold. If you want to know this for sure, wait a few years until the full restrictions kick in and all "zero emissions" cars are required to actually have zero emissions: If the GX is still sold by then, I'll concede that it is a viable product (though I'd still rather spend my money on a something like a Leaf at that price point).
If we're going to retool the entire transportation infrasturcture-- engines, pumping stations, and refineries-- I think we should aim a little higher than a minor improvement. The big problems with gasoline are the public health problem from exhaust fumes and the fact that we will run out of the stuff eventually. Natural gas doesn't solve any of those problems, and it adds transportation and potential safety problems that gasoline doesn't have.
I'd be interested in hearing from a physicist / engineer how much greater the safety problems are with natural gas. There are articles out there like this: http://www.ksl.com/?nid=148&sid=4772381 But it's not exactly fair to compare a badly done natural gas conversion with a factory-installed unit. I have a vague idea that explosive gas is bad, mmkay, but it would be interesting to hear whether it could be made safe.
Interestingly enough, I think the safest vehicle in a crash is probably a diesel vehicle, since unlike gasoline, natural gas, or batteries, a spark cannot ignite diesel fuel under normal atmospheric pressure.
As it stands, their game plan is just fine.