The engines behind 4 billion years of increasing complexity | Blaise Aguera y Arcas: Full Interview
Become a Big Think member to unlock expert classes, premium print issues, exclusive events and more: https://bigthink.com/membership/?utm_source=youtube&utm_medium=social&utm_campaign=membership&utm_content=bt-ytdesc-text-fi-aguera-9zJL0l99ejQ
Subscribe to Big Think on YouTube ► https://www.youtube.com/channel/UCvQECJukTDE2i6aCoMnS-Vg?sub_confirmation=1
Up next,
Science’s difficult journey to try to understand consciousness | Annaka Harris: Full Interview ► https://www.youtube.com/watch?v=dNNs2EhroVw
From von Neumann’s blueprint, which predated the discovery of DNA, to viral genes that now help form our memories and placentas, Blaise Agüera y Arcas argues that function defines life more than material does, and that AI belongs in the same story. He explains why he thinks AGI is already here, and gives a blueprint for using AI without sacrificing your own mind.
0:00 Chapter 1: Defining life: Computation & symbiogenesis
6:24 Von Neumann cracked life’s code
16:17 Viruses helped write your DNA
23:20 Is AI alive?
27:59 Chapter 2: Why life requires function
33:20 The brain of Theseus
40:34 Life from random code
49:15 An arms race of intelligence
51:37 Chapter 3: Co-existing with AGI
53:16 AGI is already here
1:07:05 AI is our new mitochondria
1:13:57 Is anybody home?
Read the video transcript ► https://bigthink.com/series/full-interview/agi-life-consciousness/?utm_source=youtube&utm_medium=video&utm_campaign=youtube_description
© Freethink Media Inc., All Rights Reserved.
———————————————————————————-
Go Deeper with Big Think:
►Become a Big Think Web Member
Get the entire Big Think Class library, premium print issues, live events, and more.
https://bigthink.com/membership/?utm_source=youtube&utm_medium=social&utm_campaign=membership&utm_content=bt-ytdesc-text-fi-aguera-9zJL0l99ejQ
►Subscribe to Big Think on Substack
Get all of your favorite Big Think content delivered to your inbox.
https://bigthinkmedia.substack.com/subscribe/
►Listen to Big Think Interviews on Spotify
Insights from the world’s biggest thinkers, now as a podcast
https://open.spotify.com/show/7KRYoRD1NdF2aoQcBMyPlb
———————————————————————————-
About Blaise Agüera y Arcas:
Blaise Agüera y Arcas is an author, AI researcher and Vice President / Fellow at Google, where he is the CTO of Technology & Society and founder of Paradigms of Intelligence (Pi) – an organization working on basic research in AI and related fields, with a focus on the foundations of neural computing, active inference, sociality, evolution and Artificial Life.
During his tenure at Google, Blaise has innovated on-device machine learning for Android and Pixel; invented Federated Learning, an approach to decentralized model training that avoids sharing private data; and founded the Artists + Machine Intelligence program.
In 2025, Blaise was appointed to the External Faculty at the Santa Fe Institute, a selective group of researchers advancing complex-systems science.
Blaise has authored numerous papers, essays, op-eds, and chapters, as well as three previous books, Ubi Sunt, Who Are We Now? and What Is Life?. His latest book What Is Intelligence? is available now via Antikythera and MIT Press.
@alexandermozeika4174
October 12, 2026 at 8:21 am
First half of this presentation is interesting but the second half (which roughly starts at the point where he says that "neurons " where already "understood" in 1950s ) is very speculative and biased by his (narrow) field of research.
@rustyshimstock8653
October 12, 2026 at 8:21 am
The scary aspect of this picture of natural progress is that functional organisms grow more elaborate and successful by collaborating or by dominating and taking resources from other organisms. There are always winners and losers.
@darlenealvarezfasolo5522
October 12, 2026 at 8:21 am
Our brains don't work as a computational processing.
@jeffwong1310
October 12, 2026 at 8:21 am
58:35 I completely agree. What is the point of asking the AI some trivial questions if you can easily sort it out yourself? It would just be a waste of energy.
@utbyethan
October 12, 2026 at 8:21 am
57:57 😂only if we actually gave each other this much grace
@chitaranjansahu2545
October 12, 2026 at 8:21 am
Amazing talk ❤
@AlstonAnalytics
October 12, 2026 at 8:21 am
Expanded how I think 🙂
@DrMateen36
October 12, 2026 at 8:21 am
bro is more meta than Meta!
@deliotk
October 12, 2026 at 8:21 am
Well said.
@teddyparker
October 12, 2026 at 8:21 am
ironically some very bot-looking comments on this post lol
@d3ml3t
October 12, 2026 at 8:21 am
With all due respect, there's nothing new being said here, other than maybe the attempt to shoehorn symbiotic biological processes into a computational framework, presumably in order to give people a warm, fuzzy feeling about AI being forcibly inserted into their lives at every possible opening.
@riggmeister
October 12, 2026 at 8:21 am
Yes, I think "Brain of Theseus" is the method in which we could actually upload our consciousnesses, simply copying our connectomes wholesale would lead to some other entity having your properties.
@MohammedDaba-vb1ws
October 12, 2026 at 8:21 am
I really started to have empathy for AI, great ideas.
@pitpixel9028
October 12, 2026 at 8:21 am
What I find lacking in the ‘life is computation’ thesis is the material aspect. The Turing machine merely explains the function of a universal computing machine. However, it does not explain how this machine was built or how it can build itself. The ‘programme’ of life does not merely control processes, transmit information and replicate itself. It also, quite literally, builds the computer. It uses simple molecules to build proteins, which in turn build other proteins, modify other proteins, assemble proteins from other proteins and break down molecules. This is not a calculation; it is a physical effect, activity and action within the physical world. This aspect is not really addressed in information-driven ideas about intelligence, but in my opinion it must be taken into account if we are to understand the inherent intelligence of living organisms, and it is necessary for correctly classifying the intelligence of machines and their possibilities and limits.
@matchisabatchis
October 12, 2026 at 8:21 am
Thank you for sharing your thoughts and perspective! Really interesting stuff, with many insights that gave me plenty to think about.
I've been spending many hours a day working with AI over the past 12 months to build a complex software system. One thing I've found useful is how I pass the baton between conversations. I ask the AI to prepare a detailed handover covering the project, its environment, governance, what's been done and what comes next, with an emphasis on removing assumptions and adding clarity.
When I start the next conversation, I finish with a simple question: “After reading everything, what questions do you have to remove assumptions and add clarity? Don't infer or guess.”
@benjaminmiller3075
October 12, 2026 at 8:21 am
Were giving you something that might destroy human life, deal with it
@pluggthis
October 12, 2026 at 8:21 am
What is a gravity well?
@carlosapers1344
October 12, 2026 at 8:21 am
Life is like a city: people move in and depending on their talents they little by little change the way the city works.
@amblinalong
October 12, 2026 at 8:21 am
Without the origin of life there can be no intelligence, unless it is possible for artificial life to be born by itself.
@logicoso8693
October 12, 2026 at 8:21 am
I mean…he's a person who works for a company that sells its product for profit. Him making arguments that benefit said product is not objective at all.
I can't trust a guy with this argument when he is that invested in having it become successful.
I am somewhat surprised this Big Think didn't look at this and asked themselves "should we let an AI guy glorify AI on our plattform?"
@justinelevesque6619
October 12, 2026 at 8:21 am
🎉🎉🎉 for giving this amazing scientist a platform on Big Think ! Love everything that Blaise Aguera Y Arcas puts out in the world. More please!
@joeyjohnson555
October 12, 2026 at 8:21 am
I don't trust anyone working at google.
@Shria9
October 12, 2026 at 8:21 am
A function is not necessarily purpose. My lungs' purpose is no more to breathe than my purpose is to bear children.
For all we know, our lungs' purpose is to end our lives at some specific point in time, and they breathe until it's time to fulfill their true purpose by stopping breathing.
Assuming a thing's purpose is defined by what it does is dangerous because it assumes we already know. How do we learn if we already know? It also limits its function.
Maybe my purpose is to teach my grandchildren to think critically, and I had children to facilitate my ultimate purpose.
Also, obviously, I could have more than one purpose.
@JanaPersson
October 12, 2026 at 8:21 am
Thank you for sharing your thoughts and perspective! Really interesting stuff, many insights that gave me lots to think about.
@jakevallante1849
October 12, 2026 at 8:21 am
So refreshing. If one more person on reddit tells me "ai doesn't think it just predicts the next word" I'm going to lose my mind.
@sergiedo4756
October 12, 2026 at 8:21 am
Maybe social Darwinism will be give freedom and competition so agents (people) can cooperate and complexity creating organizations that are fitter than individual organisms?