Conclusion
- Symbolic AI hasn't worked for solving the Turing test, largely
due to the symbol grounding problem.
- Neither has subsymbolic AI, largely due to structure.
- Using the brain as a guide, we can develop a grounded
symbol system.
- This is a good way to actually develop a system capable
of solving the Turing test.
- This is obviously not something that can be done over night.
- However, there is no other existing plan, that is as likely
to succeed.
