Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
Senator Schiff discusses AI regulation and free speech on The Verge - OpenSmartRoute
Senator Schiff discusses AI regulation and free speech on The Verge
Senator Adam Schiff talks about AI risks, corruption in the Trump administration, and why Congress must act now.
Key points
The Supreme Court overturned Chevron deference in Loper Bright v. Raimondo.
Tech CEOs signed a nonbinding pact to evaluate their own AI systems.
Senator Schiff chairs subcommittees on intellectual property and antitrust.
Anthropic faced pressure to stop using AI for mass surveillance.
Why it matters: Congress must regulate AI quickly because the threats are stronger than social media harms.
By OpenSmartRoute editorial · written through the router by writer-small
From The Verge AI - “Sen. Adam Schiff on AI regulation, free speech, and impeaching Trump one more time”
A stylized illustration of CA Sen. Adam Schiff’s official portrait. Image: The Verge AI (original)
The Verge podcast features Senator Adam Schiff discussing AI regulation. He talks about risks, corruption in the Trump administration, and why Congress must act now. The interview aired on October 5, 2026. Schiff sits on committees that oversee tech and AI issues. These groups handle intellectual property, antitrust, privacy, and technology laws.
Schiff asks every politician a basic question about what is happening in the country. He says things have never been seen before in recent years. There is rampant corruption within the current administration. Every day brings a new example of this problem. Campaign ads paid by taxpayer money are one major issue. The president sometimes excludes press from White House events. Other schemes involve vanity projects or questionable spending.
The Trump administration gathered tech CEOs for a voluntary safety pact. They signed a nonbinding promise to evaluate their own AI systems. There was no national security concern that made this agreement binding. Yet the companies are among the most prominent people in the country. Some of them say they might kill everyone with their technology. Schiff finds it hard to square this contradiction. The tech leaders sign pledges while saying they need regulation. They do not want the wrath of a vengeful president. They will say what he wants them to say in his presence.
The Supreme Court changed how courts treat agency regulations recently. A case called Loper Bright overturned a Reagan-era precedent. Courts used to defer to experts in agencies when the law was vague. Now the courts get to decide on these matters. This could cause any new regulation to get tied up in litigation. Schiff supports doing a new agency to regulate AI. He worries this will slow down the creation of rules.
Senator Schiff explains the challenges of regulating intellectual property. He notes that IP issues seem existential to many companies. If Congress starts charging them for training data, some might not exist anymore. The value of entertainment companies often depends on their IP. Not all players in the industry are the same size. Some have AI portfolios more important than their film libraries. Amazon's movie library is not its most important business.
AI poses new national security risks compared to past threats. The unbridled implementation of AI at the Pentagon can result in calamities. There has been an effort to go after Anthropic recently. Anthropic wanted to insist on not using AI for domestic mass surveillance. They also opposed fully autonomous weapons systems. That is a clear national security issue. The administration tried to kill Anthropic despite their stance. Lots of these concerns have never been aired publicly.
Daron Acemoglu says AI adds just 1.5 percent to global GDP over ten years. He believes human adaptation limits productivity gains more than model size.
When the president was asked about this by a reporter, he gave a weak answer. He said the world does not have to worry because companies love America. Schiff is not sure that is much of an answer. When asked if anything else is necessary, the president pointed to his brain. He claimed he has a very big brain and is a stable genius. People tell him he has passed all the tests for this role.
You are obviously on a number of committees regarding AI issues. You are the ranking member on the subcommittee on Intellectual Property. You also serve on the Privacy, Technology and the Law subcommittee. You sit on the Antitrust Competition Policy and Consumer Rights Committee too. This list places you in the dead center of the AI regulatory debate. Schiff debates these issues with colleagues like Senator Curtis. They want to know what copyrighted material was used to train models. They need companies to retain this information for discussion.
What do you think about that? Is our government prepared to start at the beginning and end at the end? Schiff says we have a holistic approach to regulation needed now. He asks if inputs are collected in a legal way that benefits the ecosystem. Studios should get paid for training data used by models. Website operators should get paid for training Google's models they scrape. Do companies need an antitrust exemption so they can coordinate on safety? They do not want to kill us all with their tech.
What is the conversation like in Congress as you debate these issues? Schiff looks at IP issues and sees them as existential to companies. If we charge for data, some businesses might cease to exist. The industry took a race-based approach to collecting input data. Companies decided to ignore IP law rather than risk falling behind. They will worry about legality later if caught. Many of these companies are now being sued by different parties. Some settle these cases while others fight them in court.
In terms of outputs, it has taken us to where we are today for these companies. They must consider the safety of the outputs now. Do we need to start slowing down the pace of model development? We do not trust the outputs enough yet. We do not know where they will lead eventually. We cannot even control these models ourselves. Providing them to good or bad actors is risky. Expecting no terrible result from public use is unrealistic.
In terms of how I think about this holistically as a legislator, this is crying out for legislation. Schiff is circumspect about Congress and legislating the tech sector. We did such a shitty job when it comes to any kind of regulation of harms. The societal harms coming out of social media were ignored for years. We must do a much better job and much more quickly with AI. The threats are even more significant now than before. We are already at the point of strong potential harms. It cannot wait another year or two.
Can I ask about that comparison between AI and social media regulation? Schiff hears it a lot from politicians, the industry, parents, and others. He believes the problem with social media regulation was never that we were too slow. The issue was always the very existence of the First Amendment. Regulating the flow of disinformation is difficult under this law. Keeping teens from bullying one another is also hard. Preventing pro-anorexia content from being posted is even harder.
All of that came down to being speech regulation in the end. Tech companies knew this would be a problem. They put up billboards in DC saying they welcome regulation. Then they crashed headfirst into the First Amendment when ideas were proposed. Nothing happened because of this legal barrier. Schiff feels like this paralyzed progress for years. We cannot rely on companies to self-regulate effectively.
The national security piece is very interesting in the context of recent questions. The Trump administration loves to justify every single thing it does by saying it is national security. This justification cannot be questioned easily. It brings more authority to the executive branch in one way or another. They have been doing this since the beginning of their term. Schiff sees this as a dangerous trend for democracy.
One nerdy note involves a concept called Chevron deference. Courts used to defer to experts in agencies when the law was vague. The Supreme Court overturned this precedent in Loper Bright. Now the courts get to decide on these matters themselves. This could cause any new regulation to get immediately tied up in litigation. Schiff supports doing a new agency to regulate AI specifically. He worries this will slow down the creation of rules significantly.
The interview has been lightly edited for length and clarity by The Verge staff. Schiff asks every politician a basic question about what is happening in the country. He says things have never been seen before in recent years. There is rampant corruption within the current administration. Every day brings a new example of this problem. Campaign ads paid by taxpayer money are one major issue.
Schiff sits on committees that oversee tech and AI issues. These groups handle intellectual property, antitrust, privacy, and technology laws. He wants to know what copyrighted material was used to train models. He needs companies to retain this information for discussion. The industry took a race-based approach to collecting input data. Companies decided to ignore IP law rather than risk falling behind.
In terms of outputs, it has taken us to where we are today for these companies. They must consider the safety of the outputs now. Do we need to start slowing down the pace of model development? We do not trust the outputs enough yet. We do not know where they will lead eventually. We cannot even control these models ourselves. Providing them to good or bad actors is risky. Expecting no terrible result from public use is unrealistic.
In terms of how I think about this holistically as a legislator, this is crying out for legislation. Schiff is circumspect about Congress and legislating the tech sector. We did such a shitty job when it comes to any kind of regulation of harms. The societal harms coming out of social media were ignored for years. We must do a much better job and much more quickly with AI. The threats are even more significant now than before. We are already at the point of strong potential harms. It cannot wait another year or two.
Can I ask about that comparison between AI and social media regulation? Schiff hears it a lot from politicians, the industry, parents, and others. He believes the problem with social media regulation was never that we were too slow. The issue was always the very existence of the First Amendment. Regulating the flow of disinformation is difficult under this law. Keeping teens from bullying one another is also hard. Preventing pro-anorexia content from being posted is even harder.
All of that came down to being speech regulation in the end. Tech companies knew this would be a problem. They put up billboards in DC saying they welcome regulation. Then they crashed headfirst into the First Amendment when ideas were proposed. Nothing happened because of this legal barrier. Schiff feels like this paralyzed progress for years. We cannot rely on companies to self-regulate effectively.
The national security piece is very interesting in the context of recent questions. The Trump administration loves to justify every single thing it does by saying it is national security. This justification cannot be questioned easily. It brings more authority to the executive branch in one way or another. They have been doing this since the beginning of their term. Schiff sees this as a dangerous trend for democracy.
One nerdy note involves a concept called Chevron deference. Courts used to defer to experts in agencies when the law was vague. The Supreme Court overturned this precedent in Loper Bright. Now the courts get to decide on these matters themselves. This could cause any new regulation to get immediately tied up in litigation. Schiff supports doing a new agency to regulate AI specifically. He worries this will slow down the creation of rules significantly.
The interview has been lightly edited for length and clarity by The Verge staff. Schiff asks every politician a basic question about what is happening in the country. He says things have never been seen before in recent years. There is rampant corruption within the current administration. Every day brings a new example of this problem. Campaign ads paid by taxpayer money are one major issue.
Schiff sits on committees that oversee tech and AI issues. These groups handle intellectual property, antitrust, privacy, and technology laws. He wants to know what copyrighted material was used to train models. He needs companies to retain this information for discussion. The industry took a race-based approach to collecting input data. Companies decided to ignore IP law rather than risk falling behind.
In terms of outputs, it has taken us to where we are today for these companies. They must consider the safety of the outputs now. Do we need to start slowing down the pace of model development? We do not trust the outputs enough yet. We do not know where they will lead eventually. We cannot even control these models ourselves. Providing them to good or bad actors is risky. Expecting no terrible result from public use is unrealistic.
In terms of how I think about this holistically as a legislator, this is crying out for legislation. Schiff is circumspect about Congress and legislating the tech sector. We did such a shitty job when it comes to any kind of regulation of harms. The societal harms coming out of social media were ignored for years. We must do a much better job and much more quickly with AI. The threats are even more significant now than before. We are already at the point of strong potential harms. It cannot wait another year or two.
Can I ask about that comparison between AI and social media regulation? Schiff hears it a lot from politicians, the industry, parents, and others. He believes the problem with social media regulation was never that we were too slow. The issue was always the very existence of the First Amendment. Regulating the flow of disinformation is difficult under this law. Keeping teens from bullying one another is also hard. Preventing pro-anorexia content from being posted is even harder.
All of that came down to being speech regulation in the end. Tech companies knew this would be a problem. They put up billboards in DC saying they welcome regulation. Then they crashed headfirst into the First Amendment when ideas were proposed. Nothing happened because of this legal barrier. Schiff feels like this paralyzed progress for years. We cannot rely on companies to self-regulate effectively.
The national security piece is very interesting in the context of recent questions. The Trump administration loves to justify every single thing it does by saying it is national security. This justification cannot be questioned easily. It brings more authority to the executive branch in one way or another. They have been doing this since the beginning of their term. Schiff sees this as a dangerous trend for democracy.
One nerdy note involves a concept called Chevron deference. Courts used to defer to experts in agencies when the law was vague. The Supreme Court overturned this precedent in Loper Bright. Now the courts get to decide on these matters themselves. This could cause any new regulation to get immediately tied up in litigation. Schiff supports doing a new agency to regulate AI specifically. He worries this will slow down the creation of rules significantly.
The interview has been lightly edited for length and clarity by The Verge staff. Schiff asks every politician a basic question about what is happening in the country. He says things have never been seen before in recent years. There is rampant corruption within the current administration. Every day brings a new example of this problem. Campaign ads paid by taxpayer money are one major issue.
Schiff sits on committees that oversee tech and AI issues. These groups handle intellectual property, antitrust, privacy, and technology laws. He wants to know what copyrighted material was used to train models. He needs companies to retain this information for discussion. The industry took a race-based approach to collecting input data. Companies decided to ignore IP law rather than risk falling behind.
In terms of outputs, it has taken us to where we are today for these companies. They must consider the safety of the outputs now. Do we need to start slowing down the pace of model development? We do not trust the outputs enough yet. We do not know where they will lead eventually. We cannot even control these models ourselves. Providing them to good or bad actors is risky. Expecting no terrible result from public use is unrealistic.
In terms of how I think about this holistically as a legislator, this is crying out for legislation. Schiff is circumspect about Congress and legislating the tech sector. We did such a shitty job when it comes to any kind of regulation of harms. The societal harms coming out of social media were ignored for years. We must do a much better job and much more quickly with AI. The threats are even more significant now than before. We are already at the point of strong potential harms. It cannot wait another year or two.
Can I ask about that comparison between AI and social media regulation? Schiff hears it a lot from politicians, the industry, parents, and others. He believes the problem with social media regulation was never that we were too slow. The issue was always the very existence of the First Amendment. Regulating the flow of disinformation is difficult under this law. Keeping teens from bullying one another is also hard. Preventing pro-anorexia content from being posted is even harder.
All of that came down to being speech regulation in the end. Tech companies knew this would be a problem. They put up billboards in DC saying they welcome regulation. Then they crashed headfirst into the First Amendment when ideas were proposed. Nothing happened because of this legal barrier. Schiff feels like this paralyzed progress for years. We cannot rely on companies to self-regulate effectively.
The national security piece is very interesting in the context of recent questions. The Trump administration loves to justify every single thing it does by saying it is national security. This justification cannot be questioned easily. It brings more authority to the executive branch in one way or another. They have been doing this since the beginning of their term. Schiff sees this as a dangerous trend for democracy.
One nerdy note involves a concept called Chevron deference. Courts used to defer to experts in agencies when the law was vague. The Supreme Court overturned this precedent in Loper Bright. Now the courts get to decide on these matters themselves. This could cause any new regulation to get immediately tied up in litigation. Schiff supports doing a new agency to regulate AI specifically. He worries this will slow down the creation of rules significantly.
The interview has been lightly edited for length and clarity by The Verge staff. Schiff asks every politician a basic question about what is happening in the country. He says things have never been seen before in recent years. There is rampant corruption within the current administration. Every day brings a new example of this problem. Campaign ads paid by taxpayer money are one major issue.
Schiff sits on committees that oversee tech and AI issues. These groups handle intellectual property, antitrust, privacy, and technology laws. He wants to know what copyrighted material was used to train models. He needs companies to retain this information for discussion. The industry took a race-based approach to collecting input data. Companies decided to ignore IP law rather than risk falling behind.
In terms of outputs, it has taken us to where we are today for these companies. They must consider the safety of the outputs now. Do we need to start slowing down the pace of model development? We do not trust the outputs enough yet. We do not know where they will lead eventually. We cannot even control these models ourselves. Providing them to good or bad actors is risky. Expecting no terrible result from public use is unrealistic.
OpenAI plans to dump hundreds of AI-solved math problems on GitHub without publishing papers. Mathematicians want formal verification and proper credit before accepting the results.
