![[Review] AI (Roman V. Yampolskiy PhD) Summarized.](https://episodes.castos.com/660078c6833215-59505987/images/2643014/c1a-085k3-xxmp5710tm58-ky4qnc.jpg)
Show Notes
AI (Roman V. Yampolskiy PhD)
- Amazon USA Store: https://www.amazon.com/dp/B0GJ1CMP9N?tag=9natree-20
- Amazon Worldwide Store: https://global.buys.trade/AI-Roman-V-Yampolskiy-PhD.html
- Apple Books: https://books.apple.com/us/audiobook/the-coming-wave-ai-power-and-our-future-unabridged/id1670283950?itsct=books_box_link&itscg=30200&ls=1&at=1001l3bAw&ct=9natree
- eBay: https://www.ebay.com/sch/i.html?_nkw=AI+Roman+V+Yampolskiy+PhD+&mkcid=1&mkrid=711-53200-19255-0&siteid=0&campid=5339060787&customid=9natree&toolid=10001&mkevt=1
- Read more: https://english.9natree.com/read/B0GJ1CMP9N/
#AIcontrolproblem #Computationalirreducibility #HaltingProblem #AIcontainment #Artificialgeneralintelligence #AI
AI: Unexplainable, Unpredictable, Uncontrollable is a 2024 nonfiction examination of artificial intelligence safety by computer scientist Roman V. Yampolskiy. Written for the Chapman and Hall CRC Artificial Intelligence and Robotics Series, it combines technical concerns about advanced computation with philosophical questions about intelligence, values, consciousness, responsibility, and human survival. Its central purpose is not to predict the next product cycle of machine learning, but to challenge the premise that a system more intelligent than its designers can reliably be understood, forecast, or kept under human control. Yampolskiy distinguishes useful, task-specific AI from artificial general intelligence and superintelligence, arguing that the latter creates a qualitatively different safety problem. The book presents the opacity of modern learning systems and theoretical limits from computer science as reasons to doubt that conventional testing, monitoring, or alignment methods can provide decisive guarantees. It is therefore a deliberately precautionary contribution to AI-risk literature, focused on what cannot be verified before irreversible deployment rather than on incremental improvements to current models.
- Amazon USA Store: https://www.amazon.com/dp/B0GJ1CMP9N?tag=9natree-20
- Amazon Worldwide Store: https://global.buys.trade/AI-Roman-V-Yampolskiy-PhD.html
- Apple Books: https://books.apple.com/us/audiobook/the-coming-wave-ai-power-and-our-future-unabridged/id1670283950?itsct=books_box_link&itscg=30200&ls=1&at=1001l3bAw&ct=9natree
- eBay: https://www.ebay.com/sch/i.html?_nkw=AI+Roman+V+Yampolskiy+PhD+&mkcid=1&mkrid=711-53200-19255-0&siteid=0&campid=5339060787&customid=9natree&toolid=10001&mkevt=1
- Read more: https://english.9natree.com/read/B0GJ1CMP9N/
#AIcontrolproblem #Computationalirreducibility #HaltingProblem #AIcontainment #Artificialgeneralintelligence #AI
AI: Unexplainable, Unpredictable, Uncontrollable is a 2024 nonfiction examination of artificial intelligence safety by computer scientist Roman V. Yampolskiy. Written for the Chapman and Hall CRC Artificial Intelligence and Robotics Series, it combines technical concerns about advanced computation with philosophical questions about intelligence, values, consciousness, responsibility, and human survival. Its central purpose is not to predict the next product cycle of machine learning, but to challenge the premise that a system more intelligent than its designers can reliably be understood, forecast, or kept under human control. Yampolskiy distinguishes useful, task-specific AI from artificial general intelligence and superintelligence, arguing that the latter creates a qualitatively different safety problem. The book presents the opacity of modern learning systems and theoretical limits from computer science as reasons to doubt that conventional testing, monitoring, or alignment methods can provide decisive guarantees. It is therefore a deliberately precautionary contribution to AI-risk literature, focused on what cannot be verified before irreversible deployment rather than on incremental improvements to current models.
