Being readable will not stop bugs. This is the history of software engineering. Heartbleed went unnoticed for years, for instance. Putting the onus on the participants to verify a bug-free smart contract is kind of absurd if you ask me. It's just not possible and definitely won't scale.
In society there are incentives for example for a bakery to not poison you. The obvious one is the financial gain, reputation, but also the threat from not following laws, regulations. There are health inspections and so on. Even with that a lot of things a bakery sells is not healthy for you.
It doesn't matter how smart the tools are, if the only incentive is to defraud people, which is how the crypto economy works today. It's all lions and hyenas, and there is no algorithm to fix that, if anything decentralized consensus makes it easier for people to obfuscate and shrug responsibility.
The same applies to blockchain in general. If a new layer-1 is released you audit the code, the token supply, distribution - if its not put out clearly to you its suspicious, otherwise you go through the same process. Those who can read first hand what the risks are can get in early, those who cannot wait. Its not rocket science.
If you cared so much about fraud victims, you would drill into new, atomic contracts and protocols, but instead you take the lazy, spiteful approach of calling the whole space a fraud and expect people to believe you because you work with computers. Can you fix my printer while you rationalize your position?