Miners need to own 51% of hash power on the network and maintain it for a while in order to do 51% attack. It is very costly if you have to hire hash power from any source and many full nodes will detect suspicious things. They will switch to other pools and the community will have solutions to stop the attack.

You are wrong because hash power is needed to run attack. Not Bitcoin. Miners need of have ASICs, and they don't need to have Bitcoin to confirm Bitcoin transactions. Same, they don't need to have Bitcoin to do 51% attack on Bitcoin network.

The Hash Power Migration since China crackdown causes a drop of more than 51% of total hash power from the peak. If any corporations, person can collect all those hash power, it would give that entity enough hash power for 51% attack.

It is impossible to collect all those ASICs into one entity. In the past, there are pools from same owners that had more than 51% hash power but they did not do any attack on the network. Because it will cause harm for Bitcoin and for everyone. It is stupid to run 51% attack.

The real threat come in when 51% of people in this world do not understand what a "51% attack" is and they post misinformation like this on a public forum. Luckily for us, more than 51% of the people in the Bitcoin community know what it is and some of us even experienced a close call, when the GHash.io mining Pool nearly acquired 51% of the hashing power.

We also know shills love to talk about "51% Attack" as if it is a huge threat to Bitcoin, but we just smile at those posts and we go on with our lives, because we know the facts.     (Done that.... got the T-Shirt.... )  

Facts : To sustain a 51% Attack is VERY expensive and definitely not profitable and if governments target 1 Crypto with that, then we just shift to the next Crypto and laugh all the way to the Bank.    (Tax money can only be wasted for a while, before the tax payer and voter starts to riot and not vote for that government)

This topic has been discussed so rigorously and the conclusion is always that no reasonable state or adversary would spend so much resources on trying to stop something so insignificant. Unless you're looking for any sorts of profits, 51% won't work. You are bound to spend a few billions in both the hardware, facilities, electricity and various other miscellaneous costs for such an insignificant attack. The money could be better used in other areas. You COULD diminish the confidence that people has in Bitcoin but another crypto would simply rise again and the cycle continues. Not to mention that ASICs are not that easy to obtain or operate in bulk.

Up until now, I thought the only way to convert fiat to BTC is through KYC exchanges. It seems like there are alternatives such as Bysk. Did I shoot myself in the foot by purchasing through KYC exchanges? 

To be honest I am certain neither argument would have calmed the anti-bitcoin party down. Whether you claim the energy is green or whether you claim it is not a lot, it doesn't fit the agenda of those against Bitcoin either way. Also, belittling an overall usage of 0.117% could be a dangerous strategy. If today you say that's not much and then it doubles within a year because Bitcoin prices go up and mining power goes up, the message in the media would be that electricity usage increased by a 100%. No matter what you say, they'll put it the way they want anyway.