I’ve been participating in misc.transport.rail.americas for a few months now, so I feel like a bit of an armchair expert on these things.

Congress has designated a number of rail routs as “high speed corridors,” which in theory should make it easier to get funding. However, no corridor has (to my knowledge) gotten any federal funding as a result of the designation. Nearly all progress in implementing the service has come from the states. Because the current administration and key congressmen are very anti-rail (such as Rep Istook), it is unlikely that any serious investment will be made until the political climate in Washington changes.
At any rate, these are the current corridors:
I believe someone said the focus was on providing 150mph service on those corridors. While that may be what Congress outlined, I believe that most of the planning has capped speeds at no more than 125mph, many at 110mph. The reason is that the FRA (Federal Railroad Administration) has designated 125mph to be the dividing line between Tier I and Tier II safety requirements. The FRA requirements are the most strict crash survivability requirements in the world. As a result, American passenger liners are the heaviest in the world. Tier II raises crash survivability to a ridiculous level. Of course, the heavier a car is, the more energy it needs to accelerate, so it’ll need a bigger plant, be less energy efficient, accelerate more slowly, and so on. Thus, American high speed rail planning assumes speeds that wouldn’t even be considered “high speed” in the rest of the world. One poster on m.t.r.a quipped that, should the corridors be built, the US would have the largest, and slowest, “high speed” rail network in the world.
Upgrading a line for high speed involves bringing the line to a state of good repair, as many of the railroads in this country are in a deplorable condition. Because they’re mostly privately owned by companies that are slowly being killed by competition with government subsidized trucking, they’re allowed to deteriorate. Rather than fix them, the railroads just put speed restrictions on them, or if they’re not using them as much, abandon them to avoid the property taxes they have to pay. The same problem exists with signaling. It’s been shown that a lot of signaling equipment currently in use dates back to the 1920s. High speed rail corridors would have their signaling equipment replaced with reliable equipment that could handle the fast moving trains. I believe, but don’t quote me on this, that 110mph is the fastest train operation the FRA will allow at grade crossings. Beyond that, the crossings must be separated. The Chicago-St. Louis corridor being planned by the State of Illinois calls for 110mph operation with full width crossing barriers that have sensors to detect if anything is stuck on the tracks. I assume those are required by the FRA for that speed but I don’t know what the lower end cutoff is. Of course, separating crossings is the best solution to the problem but also the most expensive and potentially disruptive. American trains have much higher axle loads than European trains, which means that railroad bridges (such as for grade separation), must be significantly stronger, adding still more to the cost. Finally, curves often should be eased. That is potentially very expensive and may be impossible without condemnation. It is also often one of the most important speed improvement measures, as can be seen by the poor speed of the so-called “high speed” Acela Express service between New York and Boston. One less expensive measure that may help is the increase of superelevation in the curves. Superelevation has been intentionally reduced by freight railroads throughout their networks since they use larger and slower trains to try to stay profitable.
As mentioned earlier, railroads are dying in this country. They’ve been losing market share (measured in terms of value, not tonnage) to the trucking industry for decades. The inherent energy efficiency of steel-on-steel traction can’t compete with the economic efficiency of rubber-on-taxpayer-subsidized-asphalt. Roads are all tax exempt, being owned by the government, but railroads are not. If I recall correctly, property taxes on their ROW accounts for something like 20% of all the railroads’s expenses. Railroads, as previously mentioned, are therefore keen to get rid of ROW they don’t absolutely need. They’ve been abandoning tracks for decades, and downgrading existing tracks to lower assessments. Now they’re left with heavily congested lines carrying way more traffic than that for which they were designed. Not to mention that some railroads, like UP, have decided not to run Amtrak trains on time as a matter of policy. I think one of Amtrak’s biggest problems is reliability, but that’s largely out of its hands and probably won’t be fixed unless, as another poster here suggested in another thread, the rail infrastructure is nationalized. That would be in the best interest of the railroads because it would eliminate their property tax burden, but it is questionable if they would recognize it as such. Many don’t appear to be run by logical management. Case in point is that many of the railroads use the majority of their lobbying muscle to try to attack Amtrak rather than deal with the problem of unfair, publically subsidized competition from the trucking industry. Amtrak isn’t their problem, the interstate highway system is.
As far as the Midwest goes, I think the Chicago-St. Louis and the Chicago-Milwaukee are the only two lines that have any chance at all for getting funded. Nether Michigan, Indiana, nor Ohio are very pro-rail states, and as I’ve said most of the funding will have to come from states. The Hiawatha line from Chicago-Milwaukee is one of Amtrak’s most successful lines. It also consistently has one of the best on-time performance records. High speed operation could practically bring it into the commuter rail category. The Chicago-St. Louis plan doesn’t call for any improvement north of Dwight, which will limit the usefulness of such an upgrade as that is already the second slowest section on the line (behind the Alton-St. Louis leg), and nearly all ridership on that line is between Chicago, Normal, and Springfield so most of the work will be done on tracks over which the train normally runs nearly empty.