How do we ensure that AI will have the same values as humans?
Brian Christian’s book asks questions rather than answers them. But they are good questions to ask. We have developed this powerful new tool, AI, as exemplified these days by machine learners. The more powerful it becomes, the more it impacts our daily lives. How do we ensure that it is doing what we want it to do? The alternative is that, having originated from a human design, AI learns rapidly and goes beyond where we could have foreseen. This happens all the time at the moment. There are countless stories of AI showing bias and running away with itself. The good news is that some progress is being made in designing AI whose outputs reflect our values in a more holistic way.
Link to book: https://www.waterstones.com/book/the-alignment-problem/brian-christian/9781786494337
You may also like to browse other philosophy books: https://www.thesentientrobot.com/category/philosophy/philosophy-books/