I'm not sure why WFAE refers to the Stokes county data center as a "pivot". The Matthews proposal was for 120 acres, Stokes county plan is 1800 acres and right next to the coal power plant up there. I think until we see legislation with teeth that forbids electric rate increases due to data center demands, people will continue to protest. A pledge from a company like Microsoft is essentially worthless. The difference between the data center and the factory that makes the chips - the factory provides significant numbers of jobs (in the US, for now at least). Data centers give a half dozen NOC staff a job.
Didn't this loser spend his honeymoon in the USSR? Why do we listen to this stooge. AI will have massive positive impacts. Just like Capitalism has had. But Bernie doesn't like that either. Backwards we will go!
Yep. That seems to be where America is headed. You have the Republican party thrusting us down the drain and the Democratic party organizing to thrust us down the drain. I am not positive on America's future. Short term bad, medium term really bad and long term seems to have turned to really-really bad.
- You argue AI Safety advocates & the left (incl. Bernie) haven't always worked well by default. But Bernie did platform Nate Soares, Eliezer Yudkowsky and Daniel Kokotajlo. Some of them came to our protest, giving speeches, alongside more traditional left people (eg. Sunrise)
- There also seems to be convergence on the instrumental goals: moratoriums on data centers (Bernie, datacenter groups & AI Safety people), OpenAI having less power (AI Safety, QuitGPT, child safety & anti-surveillance people), stopping the race (coalition at this protest). You don't need to agree on why to agree on what.
- You also mention a couple speakers focusing on chatbots encouraging teen suicide instead of focusing on existential risk. But if we can't stop a chatbot from killing a teenager, why would we trust these companies with superintelligence?
- Most importantly, the piece covers the protest in detail but doesn't mention that Anthropic dropped its commitment to pause development (in RSPv3) weeks before we marched on their headquarters, which was in our press release and on our website, and the reason we started at Anthropic.
Overall, I think this piece is good and I wanted to thank you again for asking great questions during your interview, and writing this very important piece: we need more detailed & balanced accounts of AI activism!
- I don't disagree that AI Safety advocates & the left have some working relationship, and it's improved over time (AI safety vs. AI ethics anyone...). Bernie in particular seems to be working to broaden the coalition. But clearly there's also been some friction
- Convergence on the instrumental goals: I agree that there are broad goals that the coalition is sympathetic to, but maybe you and I disagree how much that will translate to agreement on detailed-proposals that actually solve the important problems. For instance, I've definitely talked with some safety people who are skeptical about moratoriums on data centers because they are afraid the chips will be built in a more distributed fashion that is harder to control. There are certainly things on the margin that the whole coalition would agree with, though.
I could be wrong, but I don't think it will be easy by default.
- Yeah, agree that RSP v3 is an important part of this story. Probably an omission, but it's hard to get everything in that deserves to be in!
If you have any doubts about the threats of AI there is one simple thing you can do that will clearly show you there needs to be strict legislational limits and controls with very severe penalties for violation like natural life without pardons. Go to Utube and watch the films of the robots made by Boston Dynamic. Power these robots by AI independently or controlled from data center and there is the potential for a robot army to enslave us, kill us, eradicate our numbers. Remote control terminators being sent at citizens by tyrants. Can’t negotiate with them, can’t say I have rights, nothing. Who is going to stop the Defense Department from using such soldiers against the citizens. We need to amend constitution to stop this now!
I was very glad to read this piece, because it finally talks about the political dimensions of AI. I think the tech sector massively underestimates or simply just does not get the fact that 1. AI is already unpopular and is getting more so 2. In a democracy, you can't sustain a major technological change with most of the people opposed. And they are: 66% say AI needs to slow down according to Ipsos, people are more concerned than excited about AI with a 15 point margin according to Pew. And in qualitative surveys they are saying AI is "soulless". More then three fourths want to regulate it. Unless something changes, one or both presidential candidates are likely to run on anti-AI platform in 2028. And it does not matter that people misunderstand it (of course), or worried about the wrong things.... If you want to propose something smart about AI then yes, you have to have a meaningful coalition. But to stop something you don't need to agree on anything else besides stopping it.... The US is the country that once banned alcohol, btw,
Right... we live in a free market (more or less). The market will dictate how much we use it. In this case, however, the huge corporations are driving the AI use, so it seems the market is split between the tecnocrats and everyone else. I dont know what the answer is. I know what the answer isn't: limiting our freedoms.
It could be interesting to compare the popularity of AI and the popularity of guns. Both of them tend to be very unpopular among people who don’t use them. But people who do use them find them useful enough that they strongly oppose politicians that try to restrict it. An important difference might be that there are people who culturally support guns despite not using them themselves, and also there being enough different uses of AI that many people who use AI in some ways would strongly support limits on other ways of using it.
Yes, that would be logical, but Pew found this: "Heightened Concerns: Despite their higher engagement, young adults are more likely than their older counterparts to believe AI will have a negative impact." It is true that there is a techfan community (hello, me too:) who love AI and we might meet these opinions most often but they are numerically (meaning vote number, not net worth number:) a small minority. Although all this field is new, so caution is warranted, guess I am just saying this aspect has to enter the discussion more.
The cliche at this point is to say that (in the US) AI is unpopular but not salient. (Yet). If you ask Americans what their most important issues are, something like 0.5% say its runaway technological development. [Don't quote me on the exact number].
An unpopular, non-salient issue can be ignored politically for a while, so one of the big questions over the next couple of years is whether AI starts to become one of the main issues in the minds of voters.
Most nearly everyone opposes unlimited late-term abortions, open borders, transitioning Illegal Aliens and our children, yet that hasn’t stopped the Democrats yet.
You can’t stop AI, just because you don’t understand it, or dislike it.
Just from an environmental and economics standpoint, I wondered as I read this article what the outcry was like in Ohio, Western Pa, and Michigan with the auto and steel industry boon. I suspect people were more excited about higher paying jobs and an emerging middle class back then. As a proud Ohioan and current Pittsburgh resident, am thrilled with the economic possibilities in this region. As a human, I am terrified of its impact on humanity. We cannot stop it. We must learn how to manage it responsibly without sacrificing our global economic position.
This feels like the early formation of a trade where everyone agrees something is happening, but no one agrees on what it means.
You’ve got multiple groups circling the same asset from different angles. Some see existential risk. Some see job loss. Others just don’t want the infrastructure in their backyard. Different theses, same pressure point. That’s how coalitions try to form, but it’s also why they tend to fracture.
Markets behave the same way at inflection points. You’ll get alignment in direction without alignment in reasoning. Price starts moving, participation increases, but conviction is scattered. Everyone is in the trade for a different reason, which makes it unstable underneath.
The data center angle is interesting because it’s tangible. Water, electricity, land. Real constraints people can see and push against. That’s like liquidity in markets. When something becomes physically or financially constrained, it stops being theoretical.
What’s missing here is a unified objective. Without that, momentum tends to come in bursts rather than a sustained move. Local wins, scattered resistance, but no clean trend.
And in the meantime, the underlying system keeps building. Quietly, consistently, almost indifferent to the noise around it.
That’s usually how these things go. The debate gets louder while the infrastructure gets stronger.
I have a million thoughts on this piece. One thought that came to mind was picturing America 40 years from now akin to Cuba, with all the old cars, poverty, and stuck in another era. I feel like if we don't progress with AI, we will not be able to compete in the global economy. We will become Cuba. China and other monster economies will surpass ours with lighting speed. Our economy will be worse without AI than leveraging it and energy. If the radicals want to prohibit AI they are really asking America to become isolationist, another Iran, Cuba, or Venezuela, cutting us off from the rest of the world as they progress.
This is a profound piece. I wish there was a live forum with this where we could exchange our ideas. I spoke at 3 conferences last fall and got attacked at one by a couple of these radicals. I was just speaking on developing AI policy for education but they attacked me because my policy suggestions didn't outright ban AI because it used so much energy. Im thinking, I'm just one little school district in Western PA; do you really think I can stop this??????? People in the audience thought they were nuts, but still, they were horribly rude and you couldn't have a discussion. It was ban it or you're evil. Look, I am terrified of AI's implications on humanity, but the genie has been released. We can't put it back in the bottle now. The rest of the world will progress whether we compete or not. Not competing will destroy us for sure. Competing *may* destroy us unless we are creative with harnessing its power.
Safety protocols need to be established for sure. However, its also an existential race with the Chinese. Those who establish the AI framework and lead the technology can dominate the realm...for good or evil. The leaders who have committed mass murder of their own people and neighboring countries over the past 15 to 20 years are still in power. Our leaders whom we have disagreed with are historical figures now (meaning those still alive), sometimes expressing opinions but nothing more. So, which is worse. I would think malevolent autocratic dictatorships cannot be allowed to dominate AI. Time is ticking and they are mandating progress while we fear monger.
The political debate is necessary. But what I see in sessions is the part that doesn’t make it into bills or protests. A person sitting across from me, trying to figure out who they are now that the thing they built their life around is shifting under them. Nobody’s marching for that. There’s no moratorium on losing your sense of purpose.
Bernie’s worried about the wrong stuff rn
Most “fear of AI” is just modern technophobia—
reacting to the tool instead of understanding the system.
I'm not sure why WFAE refers to the Stokes county data center as a "pivot". The Matthews proposal was for 120 acres, Stokes county plan is 1800 acres and right next to the coal power plant up there. I think until we see legislation with teeth that forbids electric rate increases due to data center demands, people will continue to protest. A pledge from a company like Microsoft is essentially worthless. The difference between the data center and the factory that makes the chips - the factory provides significant numbers of jobs (in the US, for now at least). Data centers give a half dozen NOC staff a job.
Didn't this loser spend his honeymoon in the USSR? Why do we listen to this stooge. AI will have massive positive impacts. Just like Capitalism has had. But Bernie doesn't like that either. Backwards we will go!
Mark, yes he's a loser, in every real sense.... but (the sad part) he has thousands of people hooked onto his every losing thought...
Yep. That seems to be where America is headed. You have the Republican party thrusting us down the drain and the Democratic party organizing to thrust us down the drain. I am not positive on America's future. Short term bad, medium term really bad and long term seems to have turned to really-really bad.
If something is opposed by both Bernie Sanders and Josh Hawley, I’m more inclined to support it.
Hi Kai, Michaël Trazzi here.
Thanks for your interview and in-depth analysis.
Some comments:
- You argue AI Safety advocates & the left (incl. Bernie) haven't always worked well by default. But Bernie did platform Nate Soares, Eliezer Yudkowsky and Daniel Kokotajlo. Some of them came to our protest, giving speeches, alongside more traditional left people (eg. Sunrise)
- There also seems to be convergence on the instrumental goals: moratoriums on data centers (Bernie, datacenter groups & AI Safety people), OpenAI having less power (AI Safety, QuitGPT, child safety & anti-surveillance people), stopping the race (coalition at this protest). You don't need to agree on why to agree on what.
- You also mention a couple speakers focusing on chatbots encouraging teen suicide instead of focusing on existential risk. But if we can't stop a chatbot from killing a teenager, why would we trust these companies with superintelligence?
- Most importantly, the piece covers the protest in detail but doesn't mention that Anthropic dropped its commitment to pause development (in RSPv3) weeks before we marched on their headquarters, which was in our press release and on our website, and the reason we started at Anthropic.
Overall, I think this piece is good and I wanted to thank you again for asking great questions during your interview, and writing this very important piece: we need more detailed & balanced accounts of AI activism!
Thanks for the comments! A couple quick thoughts:
- I don't disagree that AI Safety advocates & the left have some working relationship, and it's improved over time (AI safety vs. AI ethics anyone...). Bernie in particular seems to be working to broaden the coalition. But clearly there's also been some friction
- Convergence on the instrumental goals: I agree that there are broad goals that the coalition is sympathetic to, but maybe you and I disagree how much that will translate to agreement on detailed-proposals that actually solve the important problems. For instance, I've definitely talked with some safety people who are skeptical about moratoriums on data centers because they are afraid the chips will be built in a more distributed fashion that is harder to control. There are certainly things on the margin that the whole coalition would agree with, though.
I could be wrong, but I don't think it will be easy by default.
- Yeah, agree that RSP v3 is an important part of this story. Probably an omission, but it's hard to get everything in that deserves to be in!
If you have any doubts about the threats of AI there is one simple thing you can do that will clearly show you there needs to be strict legislational limits and controls with very severe penalties for violation like natural life without pardons. Go to Utube and watch the films of the robots made by Boston Dynamic. Power these robots by AI independently or controlled from data center and there is the potential for a robot army to enslave us, kill us, eradicate our numbers. Remote control terminators being sent at citizens by tyrants. Can’t negotiate with them, can’t say I have rights, nothing. Who is going to stop the Defense Department from using such soldiers against the citizens. We need to amend constitution to stop this now!
“Bernie Sanders’s pivot toward AI safety seems like an attempt to bring these diverse forces together under one banner.”
it’s honestly been a really interesting series of event. and watching how the leaders reactive, what they target, etc. fascinating.
I was very glad to read this piece, because it finally talks about the political dimensions of AI. I think the tech sector massively underestimates or simply just does not get the fact that 1. AI is already unpopular and is getting more so 2. In a democracy, you can't sustain a major technological change with most of the people opposed. And they are: 66% say AI needs to slow down according to Ipsos, people are more concerned than excited about AI with a 15 point margin according to Pew. And in qualitative surveys they are saying AI is "soulless". More then three fourths want to regulate it. Unless something changes, one or both presidential candidates are likely to run on anti-AI platform in 2028. And it does not matter that people misunderstand it (of course), or worried about the wrong things.... If you want to propose something smart about AI then yes, you have to have a meaningful coalition. But to stop something you don't need to agree on anything else besides stopping it.... The US is the country that once banned alcohol, btw,
Right... we live in a free market (more or less). The market will dictate how much we use it. In this case, however, the huge corporations are driving the AI use, so it seems the market is split between the tecnocrats and everyone else. I dont know what the answer is. I know what the answer isn't: limiting our freedoms.
It could be interesting to compare the popularity of AI and the popularity of guns. Both of them tend to be very unpopular among people who don’t use them. But people who do use them find them useful enough that they strongly oppose politicians that try to restrict it. An important difference might be that there are people who culturally support guns despite not using them themselves, and also there being enough different uses of AI that many people who use AI in some ways would strongly support limits on other ways of using it.
Yes, that would be logical, but Pew found this: "Heightened Concerns: Despite their higher engagement, young adults are more likely than their older counterparts to believe AI will have a negative impact." It is true that there is a techfan community (hello, me too:) who love AI and we might meet these opinions most often but they are numerically (meaning vote number, not net worth number:) a small minority. Although all this field is new, so caution is warranted, guess I am just saying this aspect has to enter the discussion more.
The cliche at this point is to say that (in the US) AI is unpopular but not salient. (Yet). If you ask Americans what their most important issues are, something like 0.5% say its runaway technological development. [Don't quote me on the exact number].
An unpopular, non-salient issue can be ignored politically for a while, so one of the big questions over the next couple of years is whether AI starts to become one of the main issues in the minds of voters.
Yes!
Most nearly everyone opposes unlimited late-term abortions, open borders, transitioning Illegal Aliens and our children, yet that hasn’t stopped the Democrats yet.
You can’t stop AI, just because you don’t understand it, or dislike it.
Besides… if you try? It will get you ~;-)
Just from an environmental and economics standpoint, I wondered as I read this article what the outcry was like in Ohio, Western Pa, and Michigan with the auto and steel industry boon. I suspect people were more excited about higher paying jobs and an emerging middle class back then. As a proud Ohioan and current Pittsburgh resident, am thrilled with the economic possibilities in this region. As a human, I am terrified of its impact on humanity. We cannot stop it. We must learn how to manage it responsibly without sacrificing our global economic position.
This feels like the early formation of a trade where everyone agrees something is happening, but no one agrees on what it means.
You’ve got multiple groups circling the same asset from different angles. Some see existential risk. Some see job loss. Others just don’t want the infrastructure in their backyard. Different theses, same pressure point. That’s how coalitions try to form, but it’s also why they tend to fracture.
Markets behave the same way at inflection points. You’ll get alignment in direction without alignment in reasoning. Price starts moving, participation increases, but conviction is scattered. Everyone is in the trade for a different reason, which makes it unstable underneath.
The data center angle is interesting because it’s tangible. Water, electricity, land. Real constraints people can see and push against. That’s like liquidity in markets. When something becomes physically or financially constrained, it stops being theoretical.
What’s missing here is a unified objective. Without that, momentum tends to come in bursts rather than a sustained move. Local wins, scattered resistance, but no clean trend.
And in the meantime, the underlying system keeps building. Quietly, consistently, almost indifferent to the noise around it.
That’s usually how these things go. The debate gets louder while the infrastructure gets stronger.
I have a million thoughts on this piece. One thought that came to mind was picturing America 40 years from now akin to Cuba, with all the old cars, poverty, and stuck in another era. I feel like if we don't progress with AI, we will not be able to compete in the global economy. We will become Cuba. China and other monster economies will surpass ours with lighting speed. Our economy will be worse without AI than leveraging it and energy. If the radicals want to prohibit AI they are really asking America to become isolationist, another Iran, Cuba, or Venezuela, cutting us off from the rest of the world as they progress.
This is a profound piece. I wish there was a live forum with this where we could exchange our ideas. I spoke at 3 conferences last fall and got attacked at one by a couple of these radicals. I was just speaking on developing AI policy for education but they attacked me because my policy suggestions didn't outright ban AI because it used so much energy. Im thinking, I'm just one little school district in Western PA; do you really think I can stop this??????? People in the audience thought they were nuts, but still, they were horribly rude and you couldn't have a discussion. It was ban it or you're evil. Look, I am terrified of AI's implications on humanity, but the genie has been released. We can't put it back in the bottle now. The rest of the world will progress whether we compete or not. Not competing will destroy us for sure. Competing *may* destroy us unless we are creative with harnessing its power.
Safety protocols need to be established for sure. However, its also an existential race with the Chinese. Those who establish the AI framework and lead the technology can dominate the realm...for good or evil. The leaders who have committed mass murder of their own people and neighboring countries over the past 15 to 20 years are still in power. Our leaders whom we have disagreed with are historical figures now (meaning those still alive), sometimes expressing opinions but nothing more. So, which is worse. I would think malevolent autocratic dictatorships cannot be allowed to dominate AI. Time is ticking and they are mandating progress while we fear monger.
We are building AI systems we do not fully understand, and arguing about their risks only after they are deployed.
https://watchingatthegate.substack.com/p/is-ai-safe-no-it-is-not-safe
Two of the dumbest groups in the world:
People who think that AI execs talking about AI killing everyone are doing marketing
People who think big tech is building something that will kill everyone, but won't work with people critical of big tech
The political debate is necessary. But what I see in sessions is the part that doesn’t make it into bills or protests. A person sitting across from me, trying to figure out who they are now that the thing they built their life around is shifting under them. Nobody’s marching for that. There’s no moratorium on losing your sense of purpose.