Musk's traffic tunnel ideas are very much a pipe dream. An elevator would need to be extremely heavy-duty to lower and lift multi-ton objects quickly and would break down a lot. Getting cars into the tunnel would take too long and there would be massive bottlenecks at the entrances. A traditional underground ramp to a tunneled freeway is just the same old 20th century idea that's never worked well, and usually not at all. The only difference from Boston's Big Dig is better drilling equipment, which is not enough to make the idea workable.
Musk is very much wedded to the c1950 idea that there is some magical formula or road design where everyone can drive a gigantic metal box around, which is not surprising considering he is the CEO of a company that sells gigantic metal boxes. Most people acknowledge that traffic sucks in their city, but they blame the local government and assume there is a better way to design roads so that it does not take forever to get through a commute and find parking. The truth is that no one has found such a solution because it does not exist. Car-based transit requires more road and parking space than any city can ever build. Cities like Los Angeles have hit the limits of sprawl and are still nowhere near the point where all the traffic can flow without creating jams and bottlenecks. Attempts to remedy the issue by expanding roadways (like the recent 405 expansion) cost billions of dollars and do not reduce congestion, but only incentivize more people to drive and at peak hours.
The upshot is that auto-centric transit models only work in areas of low population density and only with massive subsidies from high-density areas because the infrastructure does not generate enough economic activity to pay for its construction and maintenance. In fact, driving in general requires massive taxpayer subsidies to work, to the tune of about $1.20 per mile. Since the average vehicle is driven about 15,000 miles per year, that's $18k per year per driver! You can trot that out the next time someone says public transit is too expensive.
The problem, of course, is the question "what's the alternative?" No one likes taking the bus, and with spread-out cities the distances are too long, and you would need an impractical number of buses and routes. The answer is bicycle infrastructure and mass transit agencies that assume people will have bikes. It is not feasible to send a bus within walking distance of everyone's house in a city like my hometown of Stockton, CA, but you could easily put a major stop or station within cycling distance (about 4 miles) and have physically separated bike lanes that can move 50 times the traffic density per lane. Even if most people keep driving, everyone who switches to bike will mean one less car causing traffic jams, adding wear and tear to the roads, and increasing the burden on the health care system, which means a nicer and cheaper place to live for everybody. People who cannot afford to drive would still be able to get around, leading to a better economy with fewer people trapped in food deserts with no practical way to get a job, which means fewer people who need government assistance. It also means fewer drunks on the road and drivers not having to share the road with slower bicycle traffic. The only people who don't benefit are auto manufacturers and oil + gas companies, which is why our governments keep pushing car-only transit models that have failed miserably for 7 decades. Personally, I don't think it's worth it for everyone to suffer a dysfunctional transit model just so that a couple of industries can make a higher profit.
I stopped reading after comments about elevator, elevators are just electric motors, if they can use them reliably on aircraft carriers then can use them for cars no problem.