One additional hypothesis I’d throw in the mix: this is a post-ZIRP adjustment.
Silicon Valley also got very comfortable with military spending after the dotcom crash.
The VC investments froze, and then post-9/11, there was a ton of government money for anyone who could promise surveillance or network analysis applications.
Then we had two decades of free investment money (along with the fallout from the Snowden revelations, as you mention.).
The free money ratcheted down in late ‘22. And, right on cue, the people who speak on behalf of Silicon Valley started talking up defense applications again.
I’m not saying it’s quite that simple, just that the ebb and flow of other sources of startup cash probably at-least-Granger-causes the rhetorical turn towards national interest arguments.
Belated response (which is all I am capable of these days) - I think that follow the money is surely a very big part, but there is also a lot of ideological action happening too. I would love to read some insider stuff or informed journalism pulling at these threads or others.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
I still have a strong suspicion bordering on belief that bitcoin was created by a US intel agency that goes by no such. Payments to spies and/or turncoats are always fraught with risk, especially with the rest of the world rapidly increasing financial surveillance capabilities. The model I have in mind is Tor - Tor by itself is useless as a method for secure comms from overseas spies, but Tor messages from spies mixed in with a witch's brew of cyber criminals, pedophiles and the privacy minded is a different story. Oh, right, and democracry advocates.
As for the Silicon Valley embrace of the national security state: I suggest watching either the shortened or the long version of the Mike Benz interview on Twitter: https://x.com/MikeBenzCyber/status/1811807660808786311
If Benz' story is the least bit accurate - and I have a very hard time disputing it - then the Silicon Valley embrace started because of the post 2014 "disinformation censorship complex" move prompted by Putin's bloodless takeover of Crimea.
From there - the vast sums of money, political access and power and general elitist nonsense created a self sustaining cycle.
But I have very little faith in this generation's Silicon Valley technologists getting diddly squat accomplished in reality, as opposed to making money. Consider the self=driving car debacle - billions spent and that stuff still isn't working. In fact, it turns out you need more engineers than cars for "self driving" cars lol. The real impact of Schmidt's White Stork (likely more aptly called White Elephant or maybe White Whale) project is the allow Western drone pilots to steer more over-expensive, under-performing wunderwaffen from Creech AFB. Silicon Valley has never shown much skill in creating real world stuff outside of Musk - why does anyone think this will change with a focus on military drones?
The whole affair seems more like the protagonists' side in Arthur C Clarke's "Superiority" short story, or a software reprise of Tiger tanks vs. the T34.
For that matter - the original Silicon Valley contributions to military were the analog and gallium arsenide and germanium, as well as silicon, control systems for various military hardware.
Has that progressed much since the 1980s? I have a hard time seeing it.
Nor is it the least bit clear that the Intels and nVidias and what not are going to be able to do much to contribute when their focus is entirely on purely civilian single digit nanometer scale chips to be used for annually replaced civilian toys like iPhones. Again, outside of Musk, where are the fantastic digital transformative technologies that have arisen in the past 30 years? Facebook? lol
As someone who worked in chip design and then later with the semiconductor manufacturing industry - the desertification of the chip design ecosystem is worse, if anything, than the desertification of the manufacturing ecosystem.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
The real issue is that defense-tech and crypto are the only two big untapped areas for venture growth. The entire tech sector has been waiting for the next iPhone and its generative AI which favors incumbents.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
If Marc Andreessen and Ben Horowitz got the memo they'd know..
• China assumed STEM leadership years ago.
• We have one research university in the world's top ten, China has seven.
• Chinese scientists's research dominate every field of science.
• 40% of America's top research papers are by ethnic Chinese.
• China added 4.7 million STEM graduates to its workforce last month.
• China outspends the USA 3:1 on R&D – $1.0 trillion to $0.3 trillion.
Pace Mark and Ben, but American technology, economic, and military leadership cannot continue for decades to come because it ended years ago. Even worse, we cannot face that simple, undeniable fact, let alone deal sanely with its consequences. Put another way, we're doomed.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
Judging by the early proliferation, it seems that producing an adequate LLM isn't super-expensive. And It seems to me (not well informed) that they are already past the point of diminishing returns. As yet, most of them seem to be cheap or free and it seems quite possible they will stay that way, with profitability to be gained only by eating into the near-monopoly of advertising revenue now held by Alphabet and Meta.
I think that both are possible - (a) we are bumping the top of the S curve, and (b) GPT4 is better than 3 and 2. And (c) that the business model is there but much more particular in scope than the massive amounts of money going into it would suggest.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
One additional hypothesis I’d throw in the mix: this is a post-ZIRP adjustment.
Silicon Valley also got very comfortable with military spending after the dotcom crash.
The VC investments froze, and then post-9/11, there was a ton of government money for anyone who could promise surveillance or network analysis applications.
Then we had two decades of free investment money (along with the fallout from the Snowden revelations, as you mention.).
The free money ratcheted down in late ‘22. And, right on cue, the people who speak on behalf of Silicon Valley started talking up defense applications again.
I’m not saying it’s quite that simple, just that the ebb and flow of other sources of startup cash probably at-least-Granger-causes the rhetorical turn towards national interest arguments.
Belated response (which is all I am capable of these days) - I think that follow the money is surely a very big part, but there is also a lot of ideological action happening too. I would love to read some insider stuff or informed journalism pulling at these threads or others.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
Sorry, but your thesis is factually wrong.
The biggest influx of VC money was not during Y2K - it was 2015.
Yes, it ratcheted down in 2022 but the absolute levels are still enormous.
The problem isn't the money, the problem is the focus or lack thereof.
Re: Crypto
I still have a strong suspicion bordering on belief that bitcoin was created by a US intel agency that goes by no such. Payments to spies and/or turncoats are always fraught with risk, especially with the rest of the world rapidly increasing financial surveillance capabilities. The model I have in mind is Tor - Tor by itself is useless as a method for secure comms from overseas spies, but Tor messages from spies mixed in with a witch's brew of cyber criminals, pedophiles and the privacy minded is a different story. Oh, right, and democracry advocates.
As for the Silicon Valley embrace of the national security state: I suggest watching either the shortened or the long version of the Mike Benz interview on Twitter: https://x.com/MikeBenzCyber/status/1811807660808786311
If Benz' story is the least bit accurate - and I have a very hard time disputing it - then the Silicon Valley embrace started because of the post 2014 "disinformation censorship complex" move prompted by Putin's bloodless takeover of Crimea.
From there - the vast sums of money, political access and power and general elitist nonsense created a self sustaining cycle.
But I have very little faith in this generation's Silicon Valley technologists getting diddly squat accomplished in reality, as opposed to making money. Consider the self=driving car debacle - billions spent and that stuff still isn't working. In fact, it turns out you need more engineers than cars for "self driving" cars lol. The real impact of Schmidt's White Stork (likely more aptly called White Elephant or maybe White Whale) project is the allow Western drone pilots to steer more over-expensive, under-performing wunderwaffen from Creech AFB. Silicon Valley has never shown much skill in creating real world stuff outside of Musk - why does anyone think this will change with a focus on military drones?
The whole affair seems more like the protagonists' side in Arthur C Clarke's "Superiority" short story, or a software reprise of Tiger tanks vs. the T34.
For that matter - the original Silicon Valley contributions to military were the analog and gallium arsenide and germanium, as well as silicon, control systems for various military hardware.
Has that progressed much since the 1980s? I have a hard time seeing it.
Nor is it the least bit clear that the Intels and nVidias and what not are going to be able to do much to contribute when their focus is entirely on purely civilian single digit nanometer scale chips to be used for annually replaced civilian toys like iPhones. Again, outside of Musk, where are the fantastic digital transformative technologies that have arisen in the past 30 years? Facebook? lol
As someone who worked in chip design and then later with the semiconductor manufacturing industry - the desertification of the chip design ecosystem is worse, if anything, than the desertification of the manufacturing ecosystem.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
The real issue is that defense-tech and crypto are the only two big untapped areas for venture growth. The entire tech sector has been waiting for the next iPhone and its generative AI which favors incumbents.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
If Marc Andreessen and Ben Horowitz got the memo they'd know..
• China assumed STEM leadership years ago.
• We have one research university in the world's top ten, China has seven.
• Chinese scientists's research dominate every field of science.
• 40% of America's top research papers are by ethnic Chinese.
• China added 4.7 million STEM graduates to its workforce last month.
• China outspends the USA 3:1 on R&D – $1.0 trillion to $0.3 trillion.
Pace Mark and Ben, but American technology, economic, and military leadership cannot continue for decades to come because it ended years ago. Even worse, we cannot face that simple, undeniable fact, let alone deal sanely with its consequences. Put another way, we're doomed.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
Their description of American military preeminence is pretty telling
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!
Judging by the early proliferation, it seems that producing an adequate LLM isn't super-expensive. And It seems to me (not well informed) that they are already past the point of diminishing returns. As yet, most of them seem to be cheap or free and it seems quite possible they will stay that way, with profitability to be gained only by eating into the near-monopoly of advertising revenue now held by Alphabet and Meta.
https://explodingtopics.com/blog/list-of-llms
I think that both are possible - (a) we are bumping the top of the S curve, and (b) GPT4 is better than 3 and 2. And (c) that the business model is there but much more particular in scope than the massive amounts of money going into it would suggest.
A general fyi which I am putting on all comments - someone seems to be going through the substack comments for this newsletter and sending messages pretending to be "programmable mutter" and suggesting contact be made on Telegram. This is presumably a phishing attack, so ignore all such messages!