Abstract
This paper outlines the non-behavioral Algorithmic Similarity criterion for machine intelligence, and assesses the likelihood that it will eventually be satisfied by computers programmed using Machine Learning (ML). Making this assessment requires overcoming the Black Box Problem, which makes it difficult to characterize the algorithms that are actually acquired via ML. This paper therefore considers Explainable AI’s prospects for solving the Black Box Problem, and for thereby providing a posteriori answers to questions about the possibility of machine intelligence. In addition, it suggests that the real-world nurture and situatedness of ML-programmed computers constitute a priori reasons for thinking that they will not only learn to behave like humans, but that they will also eventually acquire algorithms similar to the ones that are implemented in human brains.
Original language | English |
---|---|
Title of host publication | Philosophy and Theory of Artificial Intelligence 2017 |
Publisher | Springer |
Chapter | 23 |
Pages | 225-227 |
Number of pages | 3 |
ISBN (Electronic) | 978-3-319-96448-5 |
ISBN (Print) | 978-3-319-96447-8, 978-3-030-07194-3 |
DOIs | |
Publication status | Published - 2018 |
Externally published | Yes |
Event | 3rd Conference on Philosophy and Theory of Artificial Intelligence, PT-AI 2017 - Leeds, United Kingdom Duration: 4 Nov 2017 → 5 Nov 2017 Conference number: 3 |
Publication series
Name | Studies in Applied Philosophy, Epistemology and Rational Ethics (SAPERE) |
---|---|
Publisher | Springer |
Volume | 44 |
ISSN (Print) | 2192-6255 |
ISSN (Electronic) | 2192-6263 |
Conference
Conference | 3rd Conference on Philosophy and Theory of Artificial Intelligence, PT-AI 2017 |
---|---|
Country/Territory | United Kingdom |
City | Leeds |
Period | 4/11/17 → 5/11/17 |
Keywords
- Machine Learning
- Artificial Intelligence
- Turing Test