Unfortunately... People like centralized systems,-- they have lots of nice usability properties, people are willing to ignore their flaws/risks, they are easy to develop, and *critically* they are easy to monetize.
By contrast, smart contracts are essentially cryptographic protocols themselves-- inheriting the extreme difficulty and risk that normally comes with making a novel cryptographic protocol. They also inherit the development and usability challenges of distributed systems, and they're difficult or impossible to monetize-- which reduces the incentive to engage in the extraordinary amount of work needed to create them.
As a result even most things that hype "smart contracts" are pure pretext and misdirection. Mostly just borderline or outright scam tokens traded on promises which will never be met (if they're even possible to meet). Or you get insanity like Ethereum's operators going and editing the blockchain state to take funds they personally lost due to a bug in some goofy contract they were participants in and then everyone involved trying to pretend that this wasn't an example of centralization that totally mooted any purpose for a smart contract in the first place ...
But there are many things possible,
much more than even the simple example imagined by Satoshi, and the prospects seem tantalizing... but because of the lack of incentive and the risk of loss (other than for the operators of centralized systems where the funds can just be clawed back) it's a bit slow going.
Thought Bitcoin's history there have been a number of other just-for-fun smart contracting examples, e.g.
https://bitcointalksearch.org/topic/reward-offered-for-hash-collisions-for-sha1-sha256-ripemd160-and-other-293382 but a much smaller number of actual production uses.
There are some interesting smart contract things going on-- for example lightning protocol in Bitcoin is an example of using smart contracts for an actually useful purpose which isn't really just a cover for centralization or a ponzi scheme.
And most people don't even realize that it's Bitcoin smart contracts that power lightning-- maybe we should expect that when a smart contract system works people won't realize it's a smart contract at all.
Another important advance for smart contracts is, in my opinion, the
taproot proposal in Bitcoin: With taproot the common case of cooperating participants in a smart contract becomes indistinguishable from an ordinary boring payment. The full cost of public contract execution is then only needed if the contract participants try to cheat. This makes smart contract usage much more cost efficient, censorship resistant, and private. They doesn't make them less difficult to develop but it means they don't need to necessarily have a high operating cost.
There is
ongoing research going into new tools for building smart contract languages which provide for a firm theoretical grounding to them, which should make it possible for stronger analysis of smart contracts which will ultimately make it easier to author them and be confident that they'll behave like you expect.