Status Updates From The AI Mirror: How to Recla...

Ch7: regulation! AI could be useful, as long as it is transparent, contestable, and sustainable.

“If the triumph of a revolution is to be measured by the profundity of the social revisions it entrained then there has been no computer revolution”

Must get away from programs of extraction, control, domination, optimization and towards restoration, care, repair, sustenance of communities.

Ch6: another banger. AI devs are being failed: institutions crumbling and tech ethics classes don’t help. AI is not the problem… it is lack of re-assessing our virtues. Uncritical valorization of entrepreneurial/innovation virtues (efficiency, speed, seamlessness, etc) leads to bad results. Need to re-valorize and reclaim: maintenance, repair, care, service, etc. New morals needed to benefit from AI.

Ch5: Possibilities for using AI to detect patterns of bias, injustice, racism; or advance creativity. But expect it to be used to satisfy incentives: capitalism. “…the call is coming from inside the house: AI can devalue our humanity only because we already devalued it ourselves.” AI and creativity. Difference between `create` and `express`. We can make things better than what came before, this is moral creativity.

en mi interminable intento de leer teoría/no ficción. me cuesta demasiado pero este libro es muy bueno y accesible, iré dejando citas que me gustan:

“Mirrors do not only reveal us; they distort, occlude, cleave, and flatten us. And if AI is one of our most powerful mirrors today, we need to understand how its distortions and shallows dim our self-understanding and visions of our futures.”

Ch4: Algo bias and the AI threat to the space of moral reasoning. AI just has the answers… and it’s hard to defend a decision without showing the reasons. Difference between lying and bullshit. If AI crowds out human opportunity to practice morality, could lead to moral de-skilling.

Ch3: Banger! Covers existential threat mongering, EA, longtermism, Future of Humanity; counteracts utilitarianism with virtue ethics. Butler’s rise of machines (yes, the Butler of *that* Jihad) shows this is not about new tech but existing hierarchies at risk. “Superhuman” AI is more like a virus than human: relentlessly optimizing. Should be building to augment humans, not replace or turn them into meatbots.