That is absolutely not feasible right now.

To start with, many cities have no fly zones around their airports that already cut down on who you can serve. Then there's the fact that you need a sizable drone to carry a payload big enough for meals. The bigger the drone, the bigger the landing zone you'll need.

Flight times are heavily limited by batteries which means you'd have to swap batteries frequently. Then there's the fact that you'd need someone to pilot them, at least for the landing part. On top of that drones would be heavily limited by the weather.

And all that is not even getting into liability and FAA regulations. There is no way Uber or anyone else is gonna be air dropping food deliveries without charging a pretty big fee and even then there'd be a lot of limitations.

Google (Alphabet subsidiary Wing) is already operational with their drone delivery service in Australia and Finland. The drones don't land, they lower the order.

Available food payload is 1.2kg for now - which is on the "slightly limiting" for a dinner for 2 people, but many cuisines can work quite well.

> but I don't think that requires $500m in investment

Honestly, $500m sounds like a wild underestimate for what it would take to get last mile drone delivery working. Remember that this is something that's currently (mostly) illegal! There's clearly a lot of work to be done.

I don't think the fact that it's mostly illegal would discourage Uber. But compare it to Uber Eats: those driver subsidies, they're never going to get that money back. A drone last-mile R&D budget could at least lead to something. Even if it's imperfect and still requires some labor, maybe it vastly cuts cost. E.g., drivers are assigned drones, and they go from the driver's car, parked on the street of the destination, and fly to the front door to drop the food off, so drivers get more throughput while running their deliveries.