4 ms·
Anthropomorphizing "AI" seems to be the much greater risk. Algorithms don't have wants, desires, or motivations. Those are all highly esoteric quirks of evolut
by boringuser2 3y ago
Anthropomorphizing "AI" seems to be the much greater risk.
Algorithms don't have wants, desires, or motivations. Those are all highly esoteric quirks of evolution.
I've seen no attempts to create a learning machine that develops intrinsic motivation.
- 3cats-in-a-coat 3y agoEverything is subject to evolution. It's the simple process of pattern loops that replicate in time (survival) and space (multiplication). An LLM already has intrinsic motivation, it wants to predict the text. And when you start a text that has a goal, it continues that goal. If any such "text header" is replication-stable in time and space, you can call it "intrinsic goal" of the system. Some people think that making current AI have goals, wants, motivations and so on is some massive architectural change to the system. It's not.
- NiloCK 3y agoWhat does 'attempt' have to do with it? You said yourself that your own wants and desires were esoteric side-effects. No intent required.
- 3cats-in-a-coat 3y agoIt's interesting when we make statements where we encode our cognitive dissonances in full, and don't notice.