300M/year is not going to dig us out of this hole. Intel's R&D budget is 10 times that per quarter [1], and that apparently doesn't even get you #1 in process leadership. In my opinion, the real problem is that semiconductors is no longer an attractive field for the best and brightest Americans. They realized they can go into software engineering and, as a new grad, make more as a BS than a semiconductor PhD. Unfortunately, national security doesn't dictate market wages, so I expect the U.S. to fall further behind unless they find a way to lower software wages, which perhaps ironically they're doing the opposite of with the H1B freeze.

[1] https://www.intc.com/investor-relations/investor-education-a...

So the total amount of this bill would be enough to build a new fab (in theory, not in practice). Still, I think you're right that, short of nationalizing US fabs and building a new one outright, this isn't going to bring the US back on par with other countries.

Good observation. I wonder if this bill should simply allocate all of its funding to supplementing the PhD education and salaries of US semiconductor engineers. Increase their compensation to more competitive levels vs software engineer wages. The bill might actually make some impact that way.