No, that's exactly what it is.
No, it's intentional, premeditated unauthorized use of a computer of computing device. You can bet the farm that nobody authorized Burger King to assume control of their Google device and cause it to access the Internet. That it was in the form of a broadcast advertisement for a large corporation doesn't make it any less heinous to my mind.
That risk already exists, is absurdly obvious, and has been made VERY clear on both tech forums and mainstream media. Anyone with ANY clue about phone security, and this includes people who lock their phone, has already disabled the voice feature so it isn't an issue for them.
Anyone with any clue about security already avoids these things. Many with no clue about security are buying them up for the shiny factor. It is scary because the sort of people who don't have a clue are the sort of people buying them. Even people who care about security are buying and using voice activated devices.
It's about as scary as the thought that if you leave your house front door open someone could just walk in.
The average person understands that risk quite well. They might not assess its severity correctly, but they understand it. The average person does not understand the risk of these smart devices, and they remain wilfully ignorant when more knowledgable individuals try to educate them. Now we have a set of devices that can potentially be turned into a bot net en masse just by a radio or tv broadcast and the usual owner of such a device doesn't have a clue that it's even possible. Hell, they could take all the right precautions (firewall, apply updates, isolated segments, etc) and still be had. This (https://arstechnica.com/security/2017/03/smart-tv-hack-embeds-attack-code-into-broadcast-signal-no-access-required/) kind of attack springs to mind.
It's scary because Burger King will probably get away with this, paving the way for other corporates to try on the same shit.
It's scary because "OK Google" isn't necessarily the only trigger word. The attacker only needs to convince the trigger algorithm. If they discover a sound or sounds that are innocuous but trigger it then they can trigger devices without being obvious about it.
It's scary because these devices have reached a critical mass large enough that a corporation took notice and exploited them.