I don’t understand why you think DAI is centralizing. I just listened to the two latest unchained podcasts and I came to the opposite conclusion (links below). Maybe you can clarify? It would help if you provide links to published materials online that are recent as DAI has undergone a lot of changes in the last few months.
Also I think multi-collateral DAI will be even more decentralized in the future. People don’t trust stablecoins because people don’t understand stablecoins. People initially didn’t trust elevators because people didn’t understand elevators. That doesn’t mean that elevators were unsafe just that people didn’t trust them. This is why we needed full time staff to stand in elevators and press buttons and open and close elevator doors. I think we’ve come a long way as a technological society since those times. Elevators were physical objects. Stablecoins are abstract, digital, non-physical objects. Both are equally complex. Elevators could wreck your physical health. Stablecoins could wreck your financial health. But today you are not afraid of elevators are you? I think you get my point. We fear because stablecoins are new, they are complex (so were elevators), people don’t understand them (same with elevators), and they don’t have much of a track record (elevators initially so).
That’s why I’m saying the equivalent of, “hey lets build a 4 story building and put some padding at the bottom of the elevator shaft. Even if the elevator fails (which it won’t), the amount of hurt people are really going to experience will be minimal.” Doesn’t this argument sound reasonable to you? That is basically what TandaPay is. This is not the Empire State building we are talking about here. You need a different security model to go up 20+ stories. This is a 4 story elevator so please lets be reasonable. How much security do you really need to go up 4 stories?
We need to get beyond this period before we can make progress. The best way is to create apps that use DAI inconspicuously. Simply do not tell users that they are using DAI and use it. So long as the value at stake is very low and the potential loss is very minimal then the potential consequences are within the realm of manageable.
To be clear I do not think anyone should use DAI for insurance architecture where claims could potentially exceed $2,000 or even $1,000. I never said that. This would require a different security architecture which has a different security model which affords for different guarantees for the participants.
Like every complex thing in this world the true answer must be nuanced but TandaPay should never pay claims greater than $1,000 which means that DAI is the right fit given the value at stake.
If you can create models which demonstrate otherwise I’m willing to listen but if you are just speculating as to what is or is not safe then I’ll stick to my models and remain skeptical that your “intuition” is a good guide for creating real consumer financial products on the blockchain.