Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2408.08995 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929462993485824 |
|---|---|
| author | de Melo, Gabriel Adriano Maximo, Marcos Ricardo Omena De Albuquerque Soma, Nei Yoshihiro de Castro, Paulo Andre Lima |
| author_facet | de Melo, Gabriel Adriano Maximo, Marcos Ricardo Omena De Albuquerque Soma, Nei Yoshihiro de Castro, Paulo Andre Lima |
| contents | The inner alignment problem, which asserts whether an arbitrary artificial intelligence (AI) model satisfices a non-trivial alignment function of its outputs given its inputs, is undecidable. This is rigorously proved by Rice's theorem, which is also equivalent to a reduction to Turing's Halting Problem, whose proof sketch is presented in this work. Nevertheless, there is an enumerable set of provenly aligned AIs that are constructed from a finite set of provenly aligned operations. Therefore, we argue that the alignment should be a guaranteed property from the AI architecture rather than a characteristic imposed post-hoc on an arbitrary AI model. Furthermore, while the outer alignment problem is the definition of a judge function that captures human values and preferences, we propose that such a function must also impose a halting constraint that guarantees that the AI model always reaches a terminal state in finite execution steps. Our work presents examples and models that illustrate this constraint and the intricate challenges involved, advancing a compelling case for adopting an intrinsically hard-aligned approach to AI systems architectures that ensures halting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_08995 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | On the Undecidability of Artificial Intelligence Alignment: Machines that Halt de Melo, Gabriel Adriano Maximo, Marcos Ricardo Omena De Albuquerque Soma, Nei Yoshihiro de Castro, Paulo Andre Lima Artificial Intelligence The inner alignment problem, which asserts whether an arbitrary artificial intelligence (AI) model satisfices a non-trivial alignment function of its outputs given its inputs, is undecidable. This is rigorously proved by Rice's theorem, which is also equivalent to a reduction to Turing's Halting Problem, whose proof sketch is presented in this work. Nevertheless, there is an enumerable set of provenly aligned AIs that are constructed from a finite set of provenly aligned operations. Therefore, we argue that the alignment should be a guaranteed property from the AI architecture rather than a characteristic imposed post-hoc on an arbitrary AI model. Furthermore, while the outer alignment problem is the definition of a judge function that captures human values and preferences, we propose that such a function must also impose a halting constraint that guarantees that the AI model always reaches a terminal state in finite execution steps. Our work presents examples and models that illustrate this constraint and the intricate challenges involved, advancing a compelling case for adopting an intrinsically hard-aligned approach to AI systems architectures that ensures halting. |
| title | On the Undecidability of Artificial Intelligence Alignment: Machines that Halt |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2408.08995 |