blockchain is just a database

The devil is in the implementation details, but a potentially significant difference is the reduced need for trust. When the records office maintains their own database, outsiders have to trust the office to maintain and report records accurately.

Scenario 1:

You buy a house from Alice. The records office inserts a record of the transaction in their SQL Server database. You and others can confirm the transaction using a public web API the office provides.

Alice's ex-husband, Bob, has a friend in the records office. He makes a quiet payment and gets the friend to update the record, changing the seller's name from Alice to Bob. Bob then takes you to court, saying he was always the rightful owner and the sale by Alice was fraudulent. Your mortgage lender made you buy title insurance for scenarios like this, and you have a long legal battle involving paper records and hopefully reliable database backups.

Scenario 2:

The records office records the transaction in a contract on a public blockchain. Within a minute, the transaction is visible to anyone in the world, using the same tools that work for all other transactions on the blockchain. They can also review the code that makes the contract work. A bad actor could still enter a false record, but it can be immediately detected by outsiders, and valid records will be essentially immutable forever within an hour.

Any hack that takes advantage of a defect in the contract logic will put all the records at risk, and the defect will be visible to outsiders. An exploit that can rewrite the blockchain history would destroy the multibillion dollar blockchain itself.