Seems like the thesis for early impacts is that AI tech will make public spaces and platforms so shitty to exist in, a la Truth social, that everyone who can will move into private spaces and platforms, and/or that private platforms will so outcompete public platforms that the public platforms will wither.
I feel like Hammond is coming from a place of libertarian thought, like that the public sector can never make compelling products, and I question that assumption; for example, public transit is good when city design and regulation force everyone, even the rich, to use it, forming a common coalition that both funds and pushes the product towards better UX. And that does result in a better, if often less shiny, state of affairs than leaving it to private sector chaos with scooters, ubers, 3 different bus and train networks all contributing to an inefficient, unplanned, and user-unfriendly mess.
I'm unsure about the balkanization of the internet hypothesis. In the sense, the internet is already pretty oligopolistic, most of the content we see comes from the same 5-10 sites, but with enough incentives for interoperability that I wouldn't really call it balkanized. Maybe I just don't know enough about cyberthreat vectors to understand why easier cyberattack leads to nationalization of infrastructure leads to private (not public?) closed platforms.
Totally agree with the view on where open source is likely going.
Agreed that the first company to be able to run well and not go off the rails without a single human bottleneck in operations is probably going to make insane amounts of money. That said, I think that's a LOT higher bar than people expect.
"high-trust countries with ministerial systems embrace sweeping civil service reforms" - interesting thought, I bet Estonia would do well.
Hammond also says that gated industries like medicine and law will try to maintain regulatory privileges but fail, and I see where he's coming from, but idk, law is LITERALLY intertwined with government; having a black market substitute arise because it's so much cheaper is almost unthinkable. Law rests on government monopoly of force. (And sure, companies have arbitration clauses and things which selectively act as a private justice system, but they are permitted by and enforced by the regular judicial system. If you refuse to abide by the result of an arbitration, the company takes you to regular court for breach of contract, they don't put you in company jail.) For a comprehensive substitute to law to exist, you need a whole separate system of adjudication and punishment and enforcement, which would be INCREDIBLY illegal. This is kind of what organized crime has. I don't think people would voluntarily switch to a system of kneecappers just because it's cheaper and faster. Medicine, again, you need complete black market supply chains which are nevertheless trustworthy enough and free enough of corruption that grandmas will use them for heart medication. It's less absurd than law, but still a hard stretch. I get the sense that Hammond wanted to put together an scenario that gets us to a very cyberpunk future. The geofencing of LANs and effective "wild west" of the broader internet especially smacks of this, especially because there's so much obvious value to having a non-splintered internet.
"more and more administrative functions are thus offloaded onto private providers, turning the federal government into a glorified nexus of competitive contracts." - This is a real danger, a lot of the government already works like this, and I think that thankfully a lot more attention is going into govt hiring AI talent that it owns due to concerns about the private sector having all the intellectual capital in an area with rapidly escalating national security implications.
I do get that this is an illustrative scenario (and it's super fun, scary, and thought-provoking!), but I do think it's important to have some examination of assumptions, especially for a piece of writing which is likely to be taken pretty literally by a lot of people.
Our hospital is looking for a speaker to present on the topic of AI and healthcare. The context is our annual board meeting. Board includes representatives from Google Ventures, Apollo Venters, Center bridge Capital, Dell Family office...Discuss? We're in the NY metro area. selwell@wphospital.org
reading this makes me feel like it is 1995 all over again
Here in Brazil there are already closed condominiums with their own jurisdiction, I believe they are a kickstart to city-states.
Great article. Continue this trend ;)
Reading this is giving me flashbacks to the first time I read Accelerando.
Seems like the thesis for early impacts is that AI tech will make public spaces and platforms so shitty to exist in, a la Truth social, that everyone who can will move into private spaces and platforms, and/or that private platforms will so outcompete public platforms that the public platforms will wither.
I feel like Hammond is coming from a place of libertarian thought, like that the public sector can never make compelling products, and I question that assumption; for example, public transit is good when city design and regulation force everyone, even the rich, to use it, forming a common coalition that both funds and pushes the product towards better UX. And that does result in a better, if often less shiny, state of affairs than leaving it to private sector chaos with scooters, ubers, 3 different bus and train networks all contributing to an inefficient, unplanned, and user-unfriendly mess.
I'm unsure about the balkanization of the internet hypothesis. In the sense, the internet is already pretty oligopolistic, most of the content we see comes from the same 5-10 sites, but with enough incentives for interoperability that I wouldn't really call it balkanized. Maybe I just don't know enough about cyberthreat vectors to understand why easier cyberattack leads to nationalization of infrastructure leads to private (not public?) closed platforms.
Totally agree with the view on where open source is likely going.
Agreed that the first company to be able to run well and not go off the rails without a single human bottleneck in operations is probably going to make insane amounts of money. That said, I think that's a LOT higher bar than people expect.
"high-trust countries with ministerial systems embrace sweeping civil service reforms" - interesting thought, I bet Estonia would do well.
Hammond also says that gated industries like medicine and law will try to maintain regulatory privileges but fail, and I see where he's coming from, but idk, law is LITERALLY intertwined with government; having a black market substitute arise because it's so much cheaper is almost unthinkable. Law rests on government monopoly of force. (And sure, companies have arbitration clauses and things which selectively act as a private justice system, but they are permitted by and enforced by the regular judicial system. If you refuse to abide by the result of an arbitration, the company takes you to regular court for breach of contract, they don't put you in company jail.) For a comprehensive substitute to law to exist, you need a whole separate system of adjudication and punishment and enforcement, which would be INCREDIBLY illegal. This is kind of what organized crime has. I don't think people would voluntarily switch to a system of kneecappers just because it's cheaper and faster. Medicine, again, you need complete black market supply chains which are nevertheless trustworthy enough and free enough of corruption that grandmas will use them for heart medication. It's less absurd than law, but still a hard stretch. I get the sense that Hammond wanted to put together an scenario that gets us to a very cyberpunk future. The geofencing of LANs and effective "wild west" of the broader internet especially smacks of this, especially because there's so much obvious value to having a non-splintered internet.
"more and more administrative functions are thus offloaded onto private providers, turning the federal government into a glorified nexus of competitive contracts." - This is a real danger, a lot of the government already works like this, and I think that thankfully a lot more attention is going into govt hiring AI talent that it owns due to concerns about the private sector having all the intellectual capital in an area with rapidly escalating national security implications.
I do get that this is an illustrative scenario (and it's super fun, scary, and thought-provoking!), but I do think it's important to have some examination of assumptions, especially for a piece of writing which is likely to be taken pretty literally by a lot of people.
Hello Sam
Moving this to the top.
Our hospital is looking for a speaker to present on the topic of AI and healthcare. The context is our annual board meeting. Board includes representatives from Google Ventures, Apollo Venters, Center bridge Capital, Dell Family office...Discuss? We're in the NY metro area. selwell@wphospital.org
Will have to pass, sorry.
What language is this?