Abstract
Human-machine teams or systems are integral parts of society and will likely become more so. Unsettled are the effects of these changes, their mechanism(s), and how to measure them. In this article, I propose a central concept for understanding human-machine interaction: convergent cause. That is, Agent 1’s response to the object is caused by the object and Agent 2’s response, while Agent 2 responds to Agent 1’s response and the object. To the extent a human-machine team acts, AI converges with a human. One benefit of this concept is that it allows degrees, and so avoids the question of Strong or Weak AI. To defend my proposal, I repurpose Donald Davidson’s triangulation as a model for human-machine teams and systems.
Subject
Physical and Theoretical Chemistry,General Physics and Astronomy,Mathematical Physics,Materials Science (miscellaneous),Biophysics
Reference62 articles.
1. Propositional attitude reports;Nelson,2022
2. The risks of autonomy: Doyle’s catch;Woods;J Cogn Eng Decis Making,2016
3. Threat of adversarial attacks on deep learning in computer vision: A survey;Akhtar;IEEE Access,2018
4. Strike (with) A pose: Neural networks are easily fooled by strange poses of familiar objects;Alcorn,2019