AI Alignment (grant proposal)
AI Alignment (grant proposal)
Embedded Files
AI alignment research concerns itself with designing AI systems that are consistent with human values.
However, ethical discourse is typically characterised by disagreement. Moral theories disagree about which actions are right or wrong (and why), as well as which things are ultimately valuable.
But this raises a problem: How are we supposed to align AI with our values if we cannot come to agree upon what we value in the first place?
Yet for all that, there is much more that unites the different ethical theories than divides them.
Strikingly, except for a few notable exceptions, moral philosophy has focused almost exclusively on diagreement.
In this project I aim to provide a systematic account of consensus between the different ethical theories which can then serve as the basis for AI Aligment research.
Why Kant We Be Friends? AI and Human Companionship (work in progress)
Why Kant We Be Friends? AI and Human Companionship (work in progress)
A recent debate in the ethics of AI concerns the question of the value of Human-AI companionship.
In this project I offer a quasi-Kantian account of what it takes for a friend to act in ways that have intrinsic ethical worth.
I then go on to consider whether AI is capable of meeting the conditions set out in this quasi-kantian account.
I am currently revising a draft article having recently presented the paper at a number of Philosophy of AI conferences. Please email me if you are interested.
Page updated
Google Sites
Report abuse