Video Summary
Alex Bores left Palantir over ethical concerns about guardrails and immigration-related uses of the tech.
The RAISE Act (state-level) requires public safety plans, testing, incident reporting, and possible third-party audits for AI.
A proposed AI dividend would redistribute part of AI companies' gains to the public as an insurance-style safety net.
A Super PAC called Leading the Future, backed by some AI founders and investors, is bankrolling attacks to deter regulators.
Bores argues government needs more AI expertise, stronger data protections for kids, and investment in public AI infrastructure.
The Super PAC Leading the Future — supported by donors including a Palantir co-founder, some OpenAI co-founders, and prominent investors — is funding ads to oust candidates who support AI regulation, aiming to deter legislation that would constrain the industry.
The RAISE Act is a state-level AI safety bill Bores co-authored; it requires companies to publish public safety plans, routinely test models for risks, report critical safety incidents, and contemplates third-party audits and limits on unsafe model releases.
The AI dividend is a proposal to redistribute a portion of AI-generated gains back to the public as an insurance-style safety net; funding could come from taxes on capital gains, wealth taxes, or targeted taxes on AI products or corporate profits.
Bores left after internal disputes when executives declined to add guardrails he believed necessary — particularly concerns about contracts that could enable immigration enforcement and deportations.
He calls for stronger government AI capacity, public investments in safe AI research and infrastructure, data protections for children, retraining and support for displaced workers, and regulatory tools like safety plans, audits, and possible moratoriums.
"Bores did work for Palantir, but the rest of that attack is not what you might call true."
Alex Bores, a Democratic candidate in New York's 12th Congressional District, is facing relentless attacks regarding his past work with Palantir. Critics claim he profited from enabling ICE's deportations while working with the technology company.
The financing behind these attacks comes from a Super PAC named Leading the Future, which is supported by major donors, including co-founders of OpenAI and Palantir.
This coalition appears to aim at undermining candidates like Bores, who are advocating for AI regulation, making it clear that they're prioritizing corporate interests over democratic processes.
"Bores is a member of the New York State Assembly who co-authored and passed the RAISE Act, one of the first pieces of AI regulation."
Bores’s track record as a legislator is noted to be effective, successfully passing 30 bills in just over three years. Notably, he contributed to the RAISE Act, which is considered a landmark in AI regulation.
The funding of opposing Super PACs indicates a larger battle over the future of technology regulation and the political landscape surrounding AI, where entities like Leading the Future seek to stifle legislation.
The stakes are deemed high, with tech billionaires seeking to maintain their power by attempting to defeat all who might legislate against them.
"It is important that the democratic process remains more powerful than companies."
Sam Altman, co-founder of OpenAI, emphasizes the need for democratic processes to take precedence over corporate interests, especially regarding AI.
However, contradictory actions from within the AI community, such as funding anti-regulatory PACs, highlight a struggle between technological advancements and the need for oversight.
Bores's approach toward AI is characterized as balanced, indicating he does not advocate for a total rejection of AI technologies but seeks to find a rational middle ground acknowledging both risks and possibilities.
"That was my first lesson: health care needs to be universal, but also that the way we win is by working together."
Bores's political journey began in his childhood when he participated in union fights alongside his father, who was locked out by Disney during a dispute. This shaped his understanding of collective action and the importance of universal healthcare.
The exposure to grassroots activism at a young age ingrained in him the value of solidarity and organized efforts, reinforcing the idea that individual voices are often powerless without collective support.
"At the School of Industrial and Labor Relations, I learned collective bargaining and how to run campaigns that can change power and win things."
Bores studied industrial labor relations and computer science, two fields he finds increasingly intertwined. His education imbued him with knowledge of economic theory, collective bargaining, and grassroots organizing strategies.
His involvement in campaigns against corporate malfeasance, such as those against Nike's labor practices, showcased his ability to mobilize and strategically build campaigns that hold corporations accountable for their practices.
"Workers are not out there to bankrupt any company; they want the company to grow."
Bores articulates a nuanced understanding of the partnership between labor and capital, emphasizing the potential for mutual growth despite inevitable conflicts over resource distribution.
He points out that many countries, despite having favorable labor laws, struggle with enforcement, and advocates for pressure on nations that import goods to uphold labor standards.
This commentary reflects a broader discourse on the global inequities faced by workers and the systemic barriers that prevent fair labor practices.
"There was a lot of optimism that technology is going to solve some very fundamental problems of democracy."
The period around 2014 was characterized by a widespread belief that technology, particularly data systems like those offered by Palantir, would enhance government capacity while simultaneously protecting privacy and civil liberties.
This optimism led many, including employees like Alex Karp, to feel that companies like Palantir were fundamentally good and had a crucial role in delivering effective governance.
The collective mindset was that if data could be made legible and accessible, it would yield answers to complex societal problems.
"Something poisons pretty quickly... that really feels like a different ideological moment than we’re in entirely."
Post-2014, the ideology surrounding technology and governance began to change significantly, with increasing skepticism about the motives of technology companies.
This shift was punctuated by the political climate, particularly the rise of Trump in 2016, which challenged the earlier narrative of technology as a benevolent force in society.
The realization that some tech companies were aligning with questionable political regimes caused disillusionment among previous supporters.
"Fascism happens when government fails to provide for its citizens and they start blaming someone else for it."
Alex Karp's initial vision for Palantir was about safeguarding democracy against fascism by enhancing government functionality in the aftermath of 9/11.
Karp argued that a failure in governance leads to societal blame and hatred, which could spiral into fascism. The technology created was intended to make citizens feel safe without infringing on their rights.
"Very much so, the work with the banks was probably wrapping up anyway just because of time, but clearly there was no more interest in that work."
After the transition to the Trump administration, the priorities of the Department of Justice shifted drastically, which affected the nature of Palantir's work.
New requests from the DOJ included a focus on current pressing issues like the opioid epidemic and violent crime, but there was significant pushback regarding immigration-related work.
Employees at Palantir, including the speaker, felt uncomfortable and resisted involvement in projects seen as contentious or harmful, especially concerning immigration enforcement.
"I made plans to quit... executives made clear to us that they were not going to put in those guardrails."
The decision to leave Palantir emerged from internal conflicts over ethical considerations, particularly regarding contracts that could be used for deportation processes.
Despite prior focuses on non-controversial issues like human trafficking, the Trump administration pressured for expanding ICE's access to Palantir's technology for immigration enforcement.
The refusal to incorporate protective clauses against the misuse of technology for deportations contributed significantly to the speaker’s decision to resign.
"They want to beat up on me so bad that when the idea of regulating AI comes in the future, politicians run in the opposite direction."
The speaker addresses their critical stance on Palantir, revealing that executives at the company did not appreciate their comments. Despite the criticism, they received repeated requests from Palantir to return to the company after they left, indicating potential inconsistencies in the company's position on the concerns raised.
The conversation highlights the emergence of a Super PAC funded by Palantir co-founder Joe Lonsdale, which is actively working to attack the speaker due to their previous employment at Palantir. The PAC's strategy seems to be aimed at pushing back against any regulatory measures concerning AI, showcasing their fear of political accountability.
The speaker expresses concern over the lengths to which these groups will go to suppress discussions on AI regulations, indicating that their efforts are more about causing political ramifications than addressing the core issues of AI safety and ethics.
"We were seeing AI develop extremely rapidly and the industry themselves warning about what was coming."
The speaker discusses the introduction of the RAISE Act, which is one of the first significant pieces of AI legislation passed in the United States. The intent behind this act is to implement safety standards due to the rapid development of AI and the lack of federal regulations to ensure accountability and safety in the industry.
The bill requires companies to establish a public safety plan that aligns with industry best practices, ensuring that they routinely test their AI models for potential risks. Moreover, companies must document and report critical safety incidents to the government.
There were initial provisions in the bill that aimed to prevent the release of AI models if they failed safety tests, mirroring past failures of industries like tobacco and fossil fuels. Additionally, the proposal included third-party audits as a means to ensure compliance with the established safety standards, allowing for independent verification that companies are following through on their commitments.
"80 percent of Americans want to see some benefits, but see a lot of risk and think it’s moving too fast."
The speaker notes a significant shift in public sentiment regarding AI, indicating that a growing percentage of the population is now more concerned about the risks associated with AI technologies than excited about their potential. This change reflects rising anxiety about how rapidly AI is evolving.
Despite the divisive nature of many political issues today, AI remains surprisingly bipartisan, with voters across the political spectrum expressing similar concerns regarding its implications. The speaker finds this unity remarkable given the typical polarization seen in contemporary politics.
Overall, there is a notable public demand for meaningful regulation and involvement in the development of AI technologies, underscoring the need for lawmakers to address safety and ethical considerations as these systems continue to evolve and integrate into society.
"We’ve seen platforms establish with great promise, and then over time, once they get power, really turn on their users."
There is a growing skepticism regarding the narratives surrounding new technologies, particularly AI. Past technological revolutions, such as those in social media, initially offered great promise but ultimately led to negative consequences for users.
This shift in perception has made many people cautious about claims that AI will universally benefit everyone, especially when founders assert it will eliminate poverty and create material abundance.
Many individuals now doubt the likelihood of these utopian outcomes, believing instead that private companies will prioritize profit over public good.
"AI is viewed not only as a normal technology, but as an elite political project to be resisted."
The notion of AI populism describes a growing sentiment that AI is being overhyped and that its integration into society is being forced upon the public rather than embraced organically.
There is a significant investment in AI by corporations, creating a need for monetization, which compounds resistance as many feel AI is a paradigm being imposed rather than a solution to societal needs.
People express feelings of losing control over the rapid changes AI brings, generating discussions about the implications for democracy and society at large.
"There’s a sense that you could build something smart enough to disempower human beings."
While concerns about specific job losses due to automation are valid, the promises associated with AI, such as solving material scarcity, remain vague and unconvincing to the public.
Citizens are growing increasingly apprehensive about the potential harms of AI, especially when juxtaposed with the uncertain benefits it claims to offer.
This apprehension is compounded by the rapid pace at which AI is evolving, leading to fears that society may lose control over its own future.
"It becomes a question of if democratic institutions can govern this technology before it governs us."
As the pace of AI development accelerates, calls for regulation are becoming more urgent. Proposals for temporary moratoriums on data center construction have emerged as a means to pause and assess the situation.
Some advocates argue that establishing regulations is essential before further advancing AI technologies, emphasizing the need for governmental oversight.
The discussion around these moratoriums reflects a broader bipartisan consensus on the necessity to address regulatory challenges in the face of rapid technological growth.
"Whoops, that somehow our grid is still this archaic mess where the biggest problem for getting new green energy online is we can’t connect it."
The current state of the energy grid remains outdated, presenting significant barriers to integrating new green energy sources.
Despite proposals and investments, the grid struggles to connect emerging technologies and renewable energy solutions effectively.
"What if we end up in that world where all human labor is replaced, or just a significant portion of it is displaced?"
The discussion revolves around the potential of AI to replace human labor, leading to the necessity for societal adaptations such as the introduction of an AI dividend.
The AI dividend aims to provide citizens with a stake in the burgeoning AI economy, acknowledging unpredictable disruptions while planning for potential outcomes.
"The AI dividend acts almost like an insurance policy and you could fund it via boring things like a wealth tax."
Proactive measures are essential, including discussions about universal basic income (UBI) and other safety nets.
Implementing taxes on AI and capital gains can create funding channels for UBI, helping to protect those whose jobs may be displaced by AI advancements.
"The idea is not just, oh yeah, we’re all going to have this basic income; people will be screwed by that."
Relying solely on UBI won’t suffice; concurrent efforts must include job retraining programs and initiatives to support transitions into new careers.
Policymaking needs to address not just the broader economic shifts but also the individual experiences of workers affected by AI technologies, ensuring a more equitable transition.
"The real question is, should we be investing in Waymo's or in public transit?"
The rapid transition to technologies like autonomous vehicles presents challenges that have yet to be addressed adequately in policy. Suggestions to create medallions for companies like Waymo aim to create a structured transition while protecting existing jobs. However, such measures may favor medallion owners over the drivers affected by these changes.
Political leaders need to confront the complexities of technological advancements head-on and seek solutions that extend beyond binary choices of approval or rejection. Investments in retraining for jobs and a broader safety net are crucial to ensure that individuals displaced by advancements can adapt successfully.
"A world of driverless cars is safer."
While autonomous vehicles can significantly improve safety on the roads, offering solutions for individuals with mobility issues, there are considerable costs associated with the slow rollout of this technology, including safety risks and economic dislocation for drivers.
Despite the potential of driverless cars, current realities highlight limitations in their operational capabilities, raising questions about their readiness for broader deployment, especially in challenging conditions, such as rain.
"Humans have been defined by their job, and that’s become a piece of the dignity."
As AI technologies advance and automation takes over jobs, the economic implications for displaced workers cannot be ignored. Retraining may benefit individuals economically but fails to address deeper concerns regarding personal identity and societal value derived from work.
Universal Basic Income (UBI) is insufficient as a comprehensive solution, especially when considering the societal and psychological dimensions of work. The belief that future advancements will free people from the need to work needs to be scrutinized critically.
"Kids in some ways need more protection, and we don’t know a lot of the impacts that AI will have."
"Teachers and parents need a view into all of the interactions, and we need strong data protection."
The integration of AI, particularly in educational settings, raises significant concerns regarding data security and privacy. It is crucial for teachers and parents to have visibility into the interactions their students have with AI, as this can help protect sensitive information and contribute to the well-being of learners.
Mental health impacts of AI interactions should also be taken seriously, especially considering the potential influence on teenagers. There is a need for age verification and mental health checks where appropriate, which would require thoughtful implementation to avoid negative outcomes.
"Should we be requiring essays written by hand? Should we require them written in Google Docs so you can actually watch keystrokes being entered?"
The discussion around educational assessments is evolving with the rise of AI tools like ChatGPT. There is a noticeable trend where students, instead of developing their writing skills, tend to rely on AI to craft essays. This affects the assessment of their true knowledge and capabilities.
Educators might consider requiring handwritten essays or using platforms that allow for tracking of edits to ensure that students are engaging purposefully with their writing assignments. Updating pedagogical approaches to include the realities posed by AI is essential to maintain educational integrity.
"It is getting harder and harder to know what you’re looking at."
The rise of AI-generated content increasingly complicates the hiring process. Traditional aspects of evaluating writing, such as cover letters, are becoming less reliable as applicants may utilize AI tools to enhance their submissions.
There is a tension between wanting to assess an applicant's true abilities and recognizing the skill of effectively using AI. Recruiters must navigate this shift, as the competency in AI tools may not necessarily equate to genuine critical thinking or writing abilities.
"We need to be hiring in the expertise within government if we are going to help to govern and lead to good outcomes here."
Given the rapid development of AI technologies outside of traditional governmental oversight, there is a pressing need for the government to bolster its capabilities to understand and manage AI effectively.
Historically, major technological advancements have often involved government engagement, such as in the development of the internet. The current underinvestment in AI expertise within government institutions could hinder its ability to regulate and harness AI for public benefit.
"On one hand, you can’t just have a private company creating cyber super weapons and hope for the best."
The potential for AI companies to develop powerful technologies highlights significant governance challenges. The desire to avoid leaving such formidable tools solely in the hands of private enterprises is imperative, as these technologies could have substantial implications for security and control.
Proper regulatory frameworks must be established to oversee AI developments, especially as private companies engage in creating advanced systems that can function like cyber superweapons. The balance of power between private innovation and government oversight is a delicate and essential issue that needs addressing.
"I do think we should slow down the development until we’ve made a lot more progress on the alignment problem."
There is a pressing issue regarding the control of artificial intelligence, and the speaker advocates for deceleration in its development. They emphasize the need for better alignment and guarantees about what AI systems are stepping into and where they are headed.
The speaker expresses concern about entering a "really risky territory" with AI advancements, suggesting that international diplomacy and cooperation, especially with countries like China, are essential to mitigate risks associated with AI development.
The establishment of universal verification systems is proposed to monitor AI developments, both at the chip level and in the models themselves, to prevent an arms race scenario similar to the Cold War.
"I think for people to want this technology to move forward, the case has to be better than that."
The discussion shifts from harm reduction in AI technologies to the necessity of a positive vision for AI's potential benefits. The argument posits that merely focusing on reducing harm isn't enough; there needs to be a compelling vision that captures the possible good outcomes from AI innovation.
The speaker mentions the current business model for major AI companies, highlighting that they seem to rely on replacing white-collar workers and emphasizes that there ought to be a public agenda aimed at leveraging AI for applications in areas such as drug development and energy innovations.
"I absolutely believe we should be trying to use AI for good, and New York was the first state to do this."
The speaker recounts initiatives taken in New York to utilize AI for the public good, such as building the Empire AI program that provides public universities with access to GPU clusters for research and experimentation.
They highlight the importance of directing public investment toward AI research, particularly in AI alignment and safety, to ensure that developments benefit society rather than simply serving private interests.
The speaker sees potential in AI transforming drug discovery, suggesting that upcoming AI advancements could yield numerous viable drug candidates, thereby altering established processes that are currently perceived as inefficient.
"We have to cut red tape and cut regulations."
The speaker proposes that outdated regulations in New York State hinder progress, outlining efforts made to identify and remove such bureaucratic obstacles to facilitate efficient drug discovery and broader use of AI technologies.
By employing AI tools to analyze existing statutes, they aim to compile a proposal for modernizing the legal framework, thus enabling state agencies to utilize data more effectively and improve service delivery to citizens.
The conversation emphasizes the need for an agile regulatory approach to optimize AI's capabilities while ensuring safety and efficacy in its applications.
"As a public, it is a harder challenge than people have given it credit for."
Regulating AI presents significant difficulties that are often underestimated. Improving AI systems does not automatically lead to public benefits; instead, there could be both individual advantages and harms.
For industries like drug discovery to benefit from AI, there needs to be openness in the systems that facilitate this process.
There is a need to reconstruct the relationship between individuals and the state by reforming outdated government databases that are prone to errors.
"For AI to be worth it, you're going to need a lot more of this kind of investment at a much higher level of ambition."
The private sector is actively trying to adapt to AI, but the government lags behind with older systems that hinder progress.
It is crucial to increase legislative focus on AI's potential harms, but there are concerns that legislative efforts may not yield substantial benefits.
The discussion points towards the need for creating infrastructure and incentives that support the effective use of AI while ensuring thoughtful governance.
"If they really, truly believed in having one national framework that regulates AI and balances the benefits and risks, they'd be supporting me."
The speaker is running for Congress amidst opposition from powerful interests within the AI industry, including significant funding from co-founders of Palantir and OpenAI to undermine their campaign.
There is a notable disparity between what these organizations advocate for publicly versus their actual beliefs and actions regarding AI regulation.
The implementation of serious AI regulation is being obstructed by powerful entities that prefer to avoid any new frameworks that impose rules on their operations.
"At the exact moment that AI is becoming so powerful, it would be irresponsible for Congress not to start constructing regulations."
As AI technology advances, there's a recognition that Congress must take responsibility for creating regulations that can handle its potential impacts.
The amount of money the AI industry possesses gives them significant lobbying power, akin to the crypto industry's past influence, which raises concerns over the political landscape's adaptability to regulate AI effectively.
The complex political economy surrounding AI regulation poses a challenge for elected officials, who may become risk-averse due to the financial clout of the opposition.
"Politicians are terrified of the issue, and that’s the goal of leading the future."
There is a growing fear among politicians that engaging with AI regulation could jeopardize their careers, especially when faced with formidable financial opposition.
If the Super PAC opposing regulatory measures prevails, it sets a precedent that deters future legislators from advocating for necessary regulations on AI.
The implications of this scenario could lead to a political environment where elected officials are increasingly reluctant to address AI regulation, favoring superficial engagement instead of meaningful legislative action.
"So the first is my favorite book of all time, and I know you have thoughts on this book, but it’s 'A Theory of Justice' by John Rawls."
The first book recommended is "A Theory of Justice" by John Rawls, recognized as a foundational text in political philosophy. It introduces a comprehensive framework for understanding human rights and the conditions under which inequalities might be justified.
It's noted that to grasp the essentials of this book, readers should pay attention to particular sections, as Rawls specifies that the introductory portions lay the groundwork for the more complex concepts that follow.
Readers are encouraged to engage with the text more deeply to appreciate its full scope, especially considering its significance in political discourse.
"The second one is 'World Eaters' by Catherine Bracy, which is marketed as this deeply anti-vax book, but I actually think it’s written by a tech insider and takes a much more nuanced approach to the incentives that venture capital sets up."
"World Eaters" by Catherine Bracy is presented not merely as an anti-vaccine narrative, but as a critical examination of the venture capital ecosystem and its relentless push for growth without considering social ramifications.
The book highlights the detrimental impact of such growth-centric incentives, framing venture capitalists as entities that prioritize scaling a business "no matter what," leading to potential negative consequences in society.
Personal anecdotes, such as the experiences of the speaker's wife, underscore the pressures faced by entrepreneurs within the venture capital structure, illustrating the “scale or die” mentality that can affect the sustainability of businesses.
"The last one is a little more whimsical, but goes back to our conversation about the skill of writing. And it’s 'Bird by Bird' by Anne Lamott."
"Bird by Bird" by Anne Lamott is celebrated for its approachable and delightful exploration of the writing process, reminding readers, especially procrastinators, to take tasks one step at a time.
The book serves as an inspiring guide that emphasizes the importance of intentional practice in writing, especially as modern communication increasingly undermines traditional writing skills.
Lamott's advice to break down projects into manageable parts resonates well with anyone looking to improve their writing discipline, making it a valuable resource for aspiring writers.
