Marc Andreessen recently released his Techno-Optimist Manifesto, not long after James Pethokoukis released his new book The Conservative Futurist: How to Create the Sci-Fi World We Were Promised. I mostly agree with both, and the book gives much more detailed arguments. Yet, alas, I have to report that both seem to me to largely fail to inspire. I find it hard to imagine people moved to allegiance or action by these. These authors want to inspire, they try to inspire, and what they want to inspire you to is worthy of such elevation. And yet, they fail.
My interpretation of Andreessen's recent writings is that he is not trying to inspire people, per se, but instead trying to clarify how his own ideological position is different from the Democrat-Republican axis. Most people in tech are Democrats so it's particularly relevant how he differentiates from the Democratic political consensus:
1. Anti AI safety
2. Pro nuclear fission
3. Pro nuclear fusion
4. Anti ESG
5. Pro free speech
6. Anti academia
As a VC who writes large checks his job is generally not to "inspire people" - that happens earlier in the process. This is more of a "brand positioning" statement.
On inspiration. I’ve been reading a little on the origins of Critical Theory (Max Horkheimer and the Frankfurt gang), that has spread and captured the world now. Three important choices they made in the early 1930s. 1) rejected scientific positivism and even rationalism to some extent. This allowed for discussion on a more holistic level of things like fostering a new “consciousness.” Not scientifically precise but it carries a powerful meaning that resonates. It’s hard to be inspiring when the conversation centers on soulless technology and economics. And 2) while touting their socialist vision in vague holistic terms they were highly focused on hammering the cultural fissures and absurdities (there are always absurdities) of the capitalist status quo.
So maybe my suggestion is to not be soulless. What does the techno-future do to enable the thriving of humanity on a spiritual level (beyond “mere”economic prosperity). And maybe more importantly, how does the current state of stagnation and decline restrict and cripple the inner human soul. You don’t need to use the word soul or spiritual (I know that goes against the grain here) but somehow your rhetoric must accept that man does not live by bread alone.
I have an axis to grind. I prefer to keep the left-right axis (interpreting it as personal liberty versus property liberty) but add an intersecting axis that is north-south (libertarian versus authoritarian): https://jclester.substack.com/p/the-political-compass-and-why-libertarianism. I do not claim to have originated this sort of view.
Interesting take, Robin, appreciate you writing this.
Just for clarification sake, this is what I'm getting, but please correct me if I'm wrong.
Rather than arguing over what policies we're going to enact, we should focus on HOW we collect the data that informs our decision making.
Furthermore, I kind of disagree with your introduction that Marc and James' pieces on techno-optimism aren't very influential. Haven't read the Pethokoukis book yet, however, Marc's piece was definitely inspiring in my opinion. Maybe because I'm young and gullible and easy to fool, but I definitely enjoyed the overall sentiment of the piece. Sure it didn't lay out arguments with citations and such, but it definitely spurred me to go learn a bit more about Techno-Optimism and at least entertain the POV that AI won't eradicate us all.
I'd say these pieces aren't fully fleshed arguments in of themselves, but rather good starting points for people interested in these topics to dive further into.
Once again though, I do agree with your point that this endless argument of Right vs Left, Decceleration vs Acceleration, blah vs bleh is kind of pointless. Rather than endless discourse of speculation and lacking data, let's be a little bit more intentional with our time and resources rather than spitting at each other on Twitter.
Thanks Robin, appreciate you, looking forward to the next :)
Is the fact that there is not much feedback and learning involved in policy implementation considered a flaw or a feature by those who designed it? The possibility that policies might be mistaken to begin with, or need some work before they are implemented well, seems not to feature highly among those who implement them. Maybe I am being to cynical and my ignorance is showing, but it has always seemed to me that if the Donald Wittman view of the political process was close to correct, cautious learning and accumulating experience would be a much more explicit part of things.
The problem with manifestos is that they are not *arguments*. Given that in this century large tech companies have flagrantly abused their monopoly power there is understandable widespread cynicism about the tech sector amongst the general public, but Andreessen does nothing address to these concerns. I am a huger admirer of Andreessen's contributions to tech, but his recent manifesto just comes across as naive from an economic perspective (https://sphelps.substack.com/p/on-techno-optimism). For a more poetic, and well *argued*, advocacy of techno-optimism see David Deutsch's 'The Beginning of Infinity' https://www.thebeginningofinfinity.com/. From the synopsis:
"David Deutsch argues that explanations have a fundamental place in the universe. They have unlimited scope and power to cause change, and the quest to improve them is the basic regulating principle not only of science but also of all successful human endeavour. The resulting stream of ever-improving explanations has potentially infinite reach: we are subject only to the laws of physics, and they impose no upper limit to what we can eventually understand, control, and achieve."
About your final point - "how we should together aggregate info on the effects of policies" - as long as we *do* aggregate such info and use it to guide policy, that would be an improvement on the status quo. "Look at the numbers when deciding what to vote for" vs "vote how the lobbyists and talking heads tell you to vote" is already a contentious battleground. The "look at the numbers" side is often not the winning one.
But if we're dreaming of an ideal system of evaluating the outcomes of policies, I think we should and can start to build giant machine learning models of the whole economy. These models can take as input all sorts of economic data: numeric, visual (e.g. maps of economic data, satellite images), and in the form of politically relevant text (e.g. the text of bills). The models would be trained to produce predictions for the future, much like how an LLM is trained to fill in the missing word. Then, policies could be evaluated based on how positive the models say the results would be. (And models could be evaluated based on their accuracy, using standard methods of machine learning to select better and better models.)
As a scientist I am a big believer in empiricism. I've always felt one of the US's strengths is its federal system, where states retain a lot of local control and we can (and do) try out novel ideas in limited domains before scaling them up. This flexibility is in sharp contrast to our national legislature, which seems purpose-built to resist change. (E.g., the very high bar to pass Constitutional amendments. Even electing a House Speaker is an insurmountable challenge at the moment.) This combo of frothing ideas at the state level combined with the (small-c) conservatism at the national level is something akin to the id and ego of Freudian psychology. The way to optimize this is to allow maximum diversity at the state level.
Interesting, the idea of a study of what makes something inspirational is a good one. I think unfortunately having good ideas and knowing how to market those ideas are two pieces of a ven diagram that may not always intersect.
I've come to the conclusion that the axis that matters most is zero-sum vs. abundance thinking. Pethokoukis in the quote given equates zero-sum with doomer, but I think that's wrong, or rather, it's no more right than left wingers are zero-sum because a lot of right wingers are also - it literally cuts right across the traditional left-right axis.
My interpretation of Andreessen's recent writings is that he is not trying to inspire people, per se, but instead trying to clarify how his own ideological position is different from the Democrat-Republican axis. Most people in tech are Democrats so it's particularly relevant how he differentiates from the Democratic political consensus:
1. Anti AI safety
2. Pro nuclear fission
3. Pro nuclear fusion
4. Anti ESG
5. Pro free speech
6. Anti academia
As a VC who writes large checks his job is generally not to "inspire people" - that happens earlier in the process. This is more of a "brand positioning" statement.
On inspiration. I’ve been reading a little on the origins of Critical Theory (Max Horkheimer and the Frankfurt gang), that has spread and captured the world now. Three important choices they made in the early 1930s. 1) rejected scientific positivism and even rationalism to some extent. This allowed for discussion on a more holistic level of things like fostering a new “consciousness.” Not scientifically precise but it carries a powerful meaning that resonates. It’s hard to be inspiring when the conversation centers on soulless technology and economics. And 2) while touting their socialist vision in vague holistic terms they were highly focused on hammering the cultural fissures and absurdities (there are always absurdities) of the capitalist status quo.
So maybe my suggestion is to not be soulless. What does the techno-future do to enable the thriving of humanity on a spiritual level (beyond “mere”economic prosperity). And maybe more importantly, how does the current state of stagnation and decline restrict and cripple the inner human soul. You don’t need to use the word soul or spiritual (I know that goes against the grain here) but somehow your rhetoric must accept that man does not live by bread alone.
I have an axis to grind. I prefer to keep the left-right axis (interpreting it as personal liberty versus property liberty) but add an intersecting axis that is north-south (libertarian versus authoritarian): https://jclester.substack.com/p/the-political-compass-and-why-libertarianism. I do not claim to have originated this sort of view.
Consequently, I would deny that there is anything left-wing about wokeism: https://jclester.substack.com/p/wokeness-is-inverted-fascism-plus
Interesting take, Robin, appreciate you writing this.
Just for clarification sake, this is what I'm getting, but please correct me if I'm wrong.
Rather than arguing over what policies we're going to enact, we should focus on HOW we collect the data that informs our decision making.
Furthermore, I kind of disagree with your introduction that Marc and James' pieces on techno-optimism aren't very influential. Haven't read the Pethokoukis book yet, however, Marc's piece was definitely inspiring in my opinion. Maybe because I'm young and gullible and easy to fool, but I definitely enjoyed the overall sentiment of the piece. Sure it didn't lay out arguments with citations and such, but it definitely spurred me to go learn a bit more about Techno-Optimism and at least entertain the POV that AI won't eradicate us all.
I'd say these pieces aren't fully fleshed arguments in of themselves, but rather good starting points for people interested in these topics to dive further into.
Once again though, I do agree with your point that this endless argument of Right vs Left, Decceleration vs Acceleration, blah vs bleh is kind of pointless. Rather than endless discourse of speculation and lacking data, let's be a little bit more intentional with our time and resources rather than spitting at each other on Twitter.
Thanks Robin, appreciate you, looking forward to the next :)
Thanks, Robin.
There is a typo in the link in the penultimate paragraph. The correct post is https://www.overcomingbias.com/p/info-ideologyhtml
The only true progress humans have actually accomplished is our ability to live longer lives. All else are our wants.
Is the fact that there is not much feedback and learning involved in policy implementation considered a flaw or a feature by those who designed it? The possibility that policies might be mistaken to begin with, or need some work before they are implemented well, seems not to feature highly among those who implement them. Maybe I am being to cynical and my ignorance is showing, but it has always seemed to me that if the Donald Wittman view of the political process was close to correct, cautious learning and accumulating experience would be a much more explicit part of things.
The problem with manifestos is that they are not *arguments*. Given that in this century large tech companies have flagrantly abused their monopoly power there is understandable widespread cynicism about the tech sector amongst the general public, but Andreessen does nothing address to these concerns. I am a huger admirer of Andreessen's contributions to tech, but his recent manifesto just comes across as naive from an economic perspective (https://sphelps.substack.com/p/on-techno-optimism). For a more poetic, and well *argued*, advocacy of techno-optimism see David Deutsch's 'The Beginning of Infinity' https://www.thebeginningofinfinity.com/. From the synopsis:
"David Deutsch argues that explanations have a fundamental place in the universe. They have unlimited scope and power to cause change, and the quest to improve them is the basic regulating principle not only of science but also of all successful human endeavour. The resulting stream of ever-improving explanations has potentially infinite reach: we are subject only to the laws of physics, and they impose no upper limit to what we can eventually understand, control, and achieve."
Robin and Agnes Callard host a podcast? TIL! I often read Overcoming Bias but I had no idea.
About your final point - "how we should together aggregate info on the effects of policies" - as long as we *do* aggregate such info and use it to guide policy, that would be an improvement on the status quo. "Look at the numbers when deciding what to vote for" vs "vote how the lobbyists and talking heads tell you to vote" is already a contentious battleground. The "look at the numbers" side is often not the winning one.
But if we're dreaming of an ideal system of evaluating the outcomes of policies, I think we should and can start to build giant machine learning models of the whole economy. These models can take as input all sorts of economic data: numeric, visual (e.g. maps of economic data, satellite images), and in the form of politically relevant text (e.g. the text of bills). The models would be trained to produce predictions for the future, much like how an LLM is trained to fill in the missing word. Then, policies could be evaluated based on how positive the models say the results would be. (And models could be evaluated based on their accuracy, using standard methods of machine learning to select better and better models.)
Meta vs Object
Meta vs Object
As a scientist I am a big believer in empiricism. I've always felt one of the US's strengths is its federal system, where states retain a lot of local control and we can (and do) try out novel ideas in limited domains before scaling them up. This flexibility is in sharp contrast to our national legislature, which seems purpose-built to resist change. (E.g., the very high bar to pass Constitutional amendments. Even electing a House Speaker is an insurmountable challenge at the moment.) This combo of frothing ideas at the state level combined with the (small-c) conservatism at the national level is something akin to the id and ego of Freudian psychology. The way to optimize this is to allow maximum diversity at the state level.
Tribalist vs non-tribalist: tribalists value authority, hierarchy, tribal identity, conformity, adherence to dogma, strength, ingroup favoritism.
Non-tribalists value individual choice, rule of law over rule of man, critical thinking/dissent, novelty, fairness, live and let live.
Communists, right-wing authoritarians, religious fundamentalists, hard-liners of various types are examples of tribalists.
Interesting, the idea of a study of what makes something inspirational is a good one. I think unfortunately having good ideas and knowing how to market those ideas are two pieces of a ven diagram that may not always intersect.
I've come to the conclusion that the axis that matters most is zero-sum vs. abundance thinking. Pethokoukis in the quote given equates zero-sum with doomer, but I think that's wrong, or rather, it's no more right than left wingers are zero-sum because a lot of right wingers are also - it literally cuts right across the traditional left-right axis.