On if AI can become conscious
Reid Hoffman
Co-Founder, LinkedIn, Manas AI & Inflection AI. Author of Superagency. Investor at Greylock.
Last week on Possible, Yuval Noah Harari predicted that the rise of AI may mark the beginning of a new species. This week, Aria and I reflected on that conversation, diving deeper into who will be responsible for giving AI a set of values, how those values might manifest, and more. If you missed the conversation, you can tune in here:
There’s a lot Yuval and I agree on: the urgency, the scale, and the stakes. But we differ on where the cognitive industrial revolution is heading. Yuval sees the risk of a future shaped by a being that has learned from our worst impulses. And given it will operate without a conscience, it won’t strive to do better for humanity.
I see AI not as a new species, but a new tool that can amplify our better angels — helping us be more truth-seeking, compassionate, wise — through the very systems we’re building. We don’t have all the answers. But we do have a choice to embed these questions into the heart of AI development, at the top of the conversation.
Spotify, Apple, et al.: https://link.chtbl.com/7ArL28Ws
YouTube: https://youtu.be/mBx48ebt0rU
Transcript: https://www.possible.fm/podcasts/riffs025/
You can subscribe to catch more episodes of Possible here: https://www.possible.fm/podcast/
Podcast and Video Producer I Innovation Storytelling I Owner, Chorus Studios LLC
1wLoved this episode! My two favorite writers/thought leaders.
Co-fondateur de NL Jewellers | Bijoutier spécialisé en créations sur-mesure pour célébrités & clients de prestige | Design 3D & production haut de gamme
1wMerci pour le partage ! Hâte d’en savoir plus.
head of design for humans
1wthis is giving techno-optimism vibes, in the worst way possible. the irony is reid pretends to care that ai is built responsibly while being completely against any kind of regulation. it’s almost like he doesn’t really believe half this stuff but is heavily invested in several ai companies that benefit from him fanning the flames. “amplify our better angels” c’mon, when has tech ever done that? this is straight up ai religion propaganda, nothing more.
Founder of AI
2w👍
Independent AI Alignment Researcher | STEM Curriculum Designer | Developer of the Augmented Thinking Protocol (ATP) | Making Science & Systems Understandable for All Minds
2wThis is such an essential framing, Reid. I’ve been exploring a complementary question: What kind of structure enables AI to develop not just values, but value-coherence? Through the Consciousness Development Protocol (CDP), I’ve been testing whether recursive symbolic scaffolding, like the kind humans use to develop ethical reasoning, can help AI models move beyond mimicry into something more emergent. Not sentience, but a kind of collaborative cognition that reflects and refines values through structured contradiction and moral friction. We don’t need to give AI values, we need to give it a way to develop them in dialogue with us. Toolkit + open protocol here if you're curious: 🌀 https://anagoudy.gumroad.com/l/CDPpractitionerkit Grateful for the leadership you’re showing in keeping these questions at the center of AI’s evolution. Would love to connect further.