California, the epicenter of tech innovation and home to over 40 million people, often sets the tone for national policy. Its 2025 legislative session on artificial intelligence (AI) promised to address the technology’s rapid encroachment into daily life, from workplaces to children’s screens. Yet Governor Gavin Newsom’s decisions on these bills have drawn sharp criticism for favoring industry interests over public safeguards. In a series of vetoes and signings, Newsom vetoed two bills aimed at protecting workers and minors while enacting others that critics call performative at best. This outcome underscores a troubling reality: Silicon Valley’s lobbying muscle is reshaping democracy, eroding trust in institutions, and prioritizing unchecked innovation over accountability.
Tech Lobby Wins Again
The vetoes represent a clear victory for tech giants, who poured resources into opposing bills that threatened their bottom lines. SB 7, dubbed the “No Robo Bosses Act,” would have prohibited employers from relying solely on AI systems to discipline or fire workers, mandating human oversight to prevent biased or opaque automated decisions. This measure, which passed with bipartisan support, aimed to safeguard jobs amid rising AI-driven automation in hiring and performance reviews.
Similarly, AB 1064, the “LEAD Act,” sought to ensure AI chatbots were safe for minors by requiring companies to block sexual content and self-harm encouragement before market release. Newsom’s veto message cited concerns over “overly broad restrictions” that might unintentionally ban youth access to beneficial AI tools, such as educational aids. Critics, however, dismiss this as a direct echo of Silicon Valley’s playbook, where innovation is framed as untouchable to mask profit-driven risks. This rhetoric mirrors the early days of social media, when platforms lobbied against age limits and content safeguards, only to face later reckonings with youth mental health crises. By siding with industry warnings, Newsom’s actions suggest a deference to tech’s narrative that regulation stifles progress, even as evidence mounts of AI’s harms.
These vetoes did not occur in isolation. Lobbying disclosures reveal that firms like Google, Meta, and OpenAI spent millions influencing Sacramento, leveraging California’s $3.6 trillion economy and status as a tech hub to argue that such bills could drive innovation abroad. The result? Bills designed for real protections were gutted or killed, leaving workers vulnerable to algorithm-fueled layoffs and children exposed to unvetted chatbots that could exacerbate isolation or danger.
Toothless Transparency
Newsom did sign several AI-related bills, but observers argue they amount to little more than symbolic gestures, providing cover for inaction. SB 53, the “Transparency in Frontier AI Act,” targets large companies with over $500 million in revenue, requiring them to self-report safety protocols and “catastrophic risks” like events causing 50 deaths or $1 billion in damage. It also bolsters whistleblower protections, allowing employees to flag dangers without immediate retaliation.
Yet enforcement is laughably weak: violations carry fines of up to $1 million, a pittance for trillion-dollar entities like OpenAI, where such penalties barely dent quarterly profits. Whistleblower safeguards are narrow, demanding proof of imminent harm before protections kick in, which deters insiders from speaking out amid fear of non-compete clauses or blacklisting. As one policy analyst noted, this relies on the goodwill of the very companies history shows prioritize secrecy over safety.
SB 243 fares no better, mandating that AI chatbots disclose their artificial nature, prompt users on suicide and self-harm topics with resources, and suggest breaks every three hours. Non-compliance opens companies to civil suits, but penalties cap at $1,000 per incident, rendering them meaningless for deep-pocketed firms. A rare bright spot is AB 325, which regulates algorithmic dynamic pricing in housing and retail to prevent gouging, such as AI-driven rent hikes. Less contested by core tech players, it passed without major pushback, offering modest consumer relief.
These laws create an illusion of progress, with headlines touting “first-in-the-nation” reforms while doing nothing to curb AI’s real-world excesses, from biased hiring tools to addictive interfaces.
AI Regulation by Optics
The lobby’s influence is stark. Major players have mastered the art of shaping outcomes in California, where their headquarters and talent pools hold sway over politicians. Through campaign donations and astroturf coalitions, Google and Meta have blunted regulations that could require costly audits or liability, all while touting “sweeping reforms” in press releases. OpenAI, led by Sam Altman, exemplifies this: despite public pledges on safety, the company lobbied against child protections in AB 1064, framing them as barriers to “democratizing AI.” This strategy exploits California’s role as a policy bellwether, ensuring weak state laws set a low federal bar.
Nationally, the ripple effects are alarming. California’s model often inspires or preempts U.S. tech policy, so these diluted measures may stall bolder federal efforts. In Colorado, similar AI safety bills stalled amid tech opposition, mirroring Sacramento’s dynamics. The U.S. Department of Justice’s failed antitrust case against Google further entrenches monopolies, leaving citizens with scant democratic input on how AI reshapes labor, privacy, and society. Public trust erodes as voters see governance captured by unelected executives, fueling cynicism about whether democracy can tame tech’s excesses.
Where Change Can Still Happen
Looking ahead, hope emerges from grassroots momentum and cultural pushback. Anti-AI movements are surging, with Luddite-inspired campaigns demanding transparency and veto overrides, a tool unused in California since 1979 but viable with two-thirds legislative support. Figures like actress Natasha Lyonne are amplifying calls for accountability; at the TIME AI 100 event, she directly urged leaders like Altman to prioritize worker protections and ethical innovation over profit, highlighting AI’s threat to creative industries.
Ultimately, reclaiming control requires organized pressure: unions challenging AI surveillance, parents advocating for child safety, and voters demanding lobby reforms. Newsom’s choices signal Silicon Valley’s dominance, but they also ignite a broader fight for tech that serves people, not the other way around. Without it, the democratic costs of AI will only mount.