Machines that think: The obsolescence of the concept of 'strong AI'
Fuente:
Redalyc
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Artículo científico |
| Sprache: | en |
| Veröffentlicht: |
Universitat de València
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1876477161569779712 |
|---|---|
| author | Carles Fenollosa |
| author_facet | Carles Fenollosa |
| contents | Machines that think: The obsolescence of the concept of 'strong AI' Carles Fenollosa Comunicación reasoning transformers deep learning language models Artificial intelligence The development of artificial intelligence (AI) has followed a pattern of incremental advances. With the emergence of large language models, it has culminated in a turning point embodied in the launch of ChatGPT. For decades, artificial intelligence was seen as a discipline of unfulfilled promises, but deep learning has proven to be the key to developing models capable of thinking. This article reviews the evolution of AI, highlighting how current models threaten to dissolve the link between strong AI and general AI. We examine the transformer architecture, its impact on semantic representation and reasoning, and argue that these models have achieved a functional form of thinking. 2025 artículo científico 2174-3487 https://www.redalyc.org/articulo.oa?id=511783270002 https://www.redalyc.org/journal/5117/511783270002/ https://www.redalyc.org/journal/5117/511783270002/html/ https://www.redalyc.org/journal/5117/511783270002/511783270002.epub https://www.redalyc.org/journal/5117/511783270002/movil 10.7203/metode.15.30523 en http://www.redalyc.org/revista.oa?id=5117 Mètode Science Studies Journal application/pdf Universitat de València Mètode Science Studies Journal (España) Num.7 Vol.15 |
| format | Artículo científico |
| id | redalyc_511783270002 |
| institution | Redalyc |
| language | en |
| publishDate | 2025 |
| publisher | Universitat de València |
| spellingShingle | Machines that think: The obsolescence of the concept of 'strong AI' Carles Fenollosa Comunicación reasoning transformers deep learning language models Artificial intelligence Machines that think: The obsolescence of the concept of 'strong AI' Carles Fenollosa Comunicación reasoning transformers deep learning language models Artificial intelligence The development of artificial intelligence (AI) has followed a pattern of incremental advances. With the emergence of large language models, it has culminated in a turning point embodied in the launch of ChatGPT. For decades, artificial intelligence was seen as a discipline of unfulfilled promises, but deep learning has proven to be the key to developing models capable of thinking. This article reviews the evolution of AI, highlighting how current models threaten to dissolve the link between strong AI and general AI. We examine the transformer architecture, its impact on semantic representation and reasoning, and argue that these models have achieved a functional form of thinking. 2025 artículo científico 2174-3487 https://www.redalyc.org/articulo.oa?id=511783270002 https://www.redalyc.org/journal/5117/511783270002/ https://www.redalyc.org/journal/5117/511783270002/html/ https://www.redalyc.org/journal/5117/511783270002/511783270002.epub https://www.redalyc.org/journal/5117/511783270002/movil 10.7203/metode.15.30523 en http://www.redalyc.org/revista.oa?id=5117 Mètode Science Studies Journal application/pdf Universitat de València Mètode Science Studies Journal (España) Num.7 Vol.15 |
| title | Machines that think: The obsolescence of the concept of 'strong AI' |
| topic | Comunicación reasoning transformers deep learning language models Artificial intelligence |
| url | https://www.redalyc.org/articulo.oa?id=511783270002 https://www.redalyc.org/journal/5117/511783270002/ https://www.redalyc.org/journal/5117/511783270002/html/ https://www.redalyc.org/journal/5117/511783270002/511783270002.epub https://www.redalyc.org/journal/5117/511783270002/movil |