Robust AI Security and Alignment: A Sisyphean Endeavor?
Fuente:
arXiv
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910223013249024 |
|---|---|
| author | Vassilev, Apostol |
| author_facet | Vassilev, Apostol |
| contents | This manuscript establishes information-theoretic limitations for robustness of AI security and alignment by extending Gödel's incompleteness theorem to AI. Knowing these limitations and preparing for the challenges they bring is critically important for the responsible adoption of the AI technology. Practical approaches to dealing with these challenges are provided as well. Broader implications for cognitive reasoning limitations of AI systems are also proven. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_10100 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Robust AI Security and Alignment: A Sisyphean Endeavor? Vassilev, Apostol Artificial Intelligence This manuscript establishes information-theoretic limitations for robustness of AI security and alignment by extending Gödel's incompleteness theorem to AI. Knowing these limitations and preparing for the challenges they bring is critically important for the responsible adoption of the AI technology. Practical approaches to dealing with these challenges are provided as well. Broader implications for cognitive reasoning limitations of AI systems are also proven. |
| title | Robust AI Security and Alignment: A Sisyphean Endeavor? |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2512.10100 |