#bostrum

Tom recommended I look into Nick Bostrum for some philosophical context. At first, most of what appeared was dominated by AI and its ethical issues so i struggled to see the relevance but then after watching a short talk, I noticed a parallel between Bostrums preventative measures for controlling superintelligence in AI and my work. 

Bostrums proposes that one solution to the problem of ‘runaway superintelligence’ could be to structure future AI to learn to share our human values, so as to prevent it sumising that humans are a threat to any of its programmed objectives. 

So where I am discussing the transfer of values from one ideological context (capitalism, neoliberalism, etc.) to another (the ideal ‘sustainable’ future*), Bostrum is talking about the transfer of values between human > non human intelligence. 

Bostrum also nods to the myth of King Midas and his golden touch. Even in a conversation about AI - one of the most infamous futuristic concepts - we can't help but refer back to myth and lore. Myths are such a focal point for how we taught/teach morality and value systems that I want to look into them further. 

With regards to Bostrums other work, i'm unsure as to the extent which it is relevant. Apart from surface parallels, I feel the work is pulling me in a more philosophical/existential direction than I want to take. I want to remain in the practical/pragmatic sphere and to try to act and think more in the present about what is required in the short-term future in the impending transitionary period.

*This in itself would need clarifying, for a quick definition, to me it means a future ideology that does not function at the expense of select groups of participants, or the physical environment it exists within. Further exploration of the ‘ideal’ scenario is definitely needed. 

Links:

https://www.youtube.com/watch?v=MnT1xgZgkpk