Real Sparks of Artificial Intelligence and the Importance of Inner Interpretability

Inquiry: An Interdisciplinary Journal of Philosophy (forthcoming)
  Copy   BIBTEX

Abstract

The present paper looks at one of the most thorough articles on the intelligence of GPT, research conducted by engineers at Microsoft. Although there is a great deal of value in their work, I will argue that, for familiar philosophical reasons, their methodology, ‘Black-box Interpretability’ is wrongheaded. But there is a better way. There is an exciting and emerging discipline of ‘Inner Interpretability’ (also sometimes called ‘White-box Interpretability’) that aims to uncover the internal activations and weights of models in order to understand what they represent and the algorithms they implement. In my view, a crucial mistake in Black-box Interpretability is the failure to appreciate that how processes are carried out matters when it comes to intelligence and understanding. I can’t pretend to have a full story that provides both necessary and sufficient conditions for being intelligent, but I do think that Inner Interpretability dovetails nicely with plausible philosophical views of what intelligence requires. So the conclusion is modest, but the important point in my view is seeing how to get the research on the right track. Towards the end of the paper, I will show how some of the philosophical concepts can be used to further refine how Inner Interpretability is approached, so the paper helps draw out a profitable, future two-way exchange between Philosophers and Computer Scientists.

Similar books and articles

Complexity of the interpretability logics ILW and ILP.Luka Mikec - 2023 - Logic Journal of the IGPL 31 (1):194-213.
What is Interpretability?Adrian Erasmus, Tyler D. P. Brunet & Eyal Fisher - 2021 - Philosophy and Technology 34:833–862.
Interpretability over peano arithmetic.Claes Strannegård - 1999 - Journal of Symbolic Logic 64 (4):1407-1425.
The formalization of interpretability.Albert Visser - 1991 - Studia Logica 50 (1):81 - 105.
Interpretability over peano arithmetic.Claes Strannegård - 1999 - Journal of Symbolic Logic 64 (4):1407-1425.
The virtues of interpretable medical artificial intelligence.Joshua Hatherley, Robert Sparrow & Mark Howard - forthcoming - Cambridge Quarterly of Healthcare Ethics:1-10.
Interpretability over peano arithmetic.Claes Strannegård - 1999 - Journal of Symbolic Logic 64 (4):1407-1425.
Interpolation and the Interpretability Logic of PA.Evan Goris - 2006 - Notre Dame Journal of Formal Logic 47 (2):179-195.
Interpretability in.Marta Bílková, Dick de Jongh & Joost J. Joosten - 2010 - Annals of Pure and Applied Logic 161 (2):128-138.
On Interpretability of Almost Linear Orderings.Akito Tsuboi & Kentaro Wakai - 1998 - Notre Dame Journal of Formal Logic 39 (3):325-331.

Analytics

Added to PP
2023-09-11

Downloads
374 (#51,469)

6 months
238 (#9,692)

Historical graph of downloads
How can I increase my downloads?

Author's Profile

Alex Grzankowski
Birkbeck, University of London

Citations of this work

No citations found.

Add more citations