Desiree Sainthrope brings a seasoned perspective to the high-stakes arena of technology law, having spent years dissecting the complexities of trade agreements and global compliance. Her expertise sits at the intersection of emerging technologies and legislative architecture, making her a vital voice in understanding how the legal landscape shifts when innovation outpaces regulation. As statehouses across the country move from cautious observation to aggressive intervention, Sainthrope’s deep understanding of intellectual property and AI policy provides a window into a historic power struggle between Silicon Valley and local lawmakers.
The following discussion explores the dramatic erosion of tech industry influence within state legislatures, where a once-feared lobby now finds its traditional tactics—such as the threat of a “patchwork” of conflicting laws—falling on deaf ears. We delve into the surge of pro-regulation funding that has leveled the playing field, the profound shift in public sentiment that has turned AI oversight into a bipartisan rallying cry, and the strategic pivot of major tech players who are now scrambling to shape the very rules they once tried to block. From the failure of federal “paper tiger” threats to the role of governors as the final line of defense for industry interests, this conversation maps the new geography of American AI regulation.
State lawmakers are increasingly advancing safety bills and auditing standards despite warnings about an unworkable “patchwork” of conflicting regulations. From your perspective, why has this specific industry warning—which used to be a silver bullet for stopping state action—lost its effectiveness in the current legislative environment?
The “patchwork” argument has essentially hit a wall because legislators have stopped viewing it as a logistical concern and started seeing it as a delay tactic used to preserve a regulatory vacuum. In the past, the mere threat of fifty different sets of rules was enough to freeze a statehouse in its tracks, but that fear has been replaced by a much more visceral concern: the safety of their own constituents. When I look at states like Illinois, where they recently passed a law requiring top AI developers to submit to independent audits of their safety plans, it’s clear that the “train is moving,” as many local leaders like to say. The industry tried to frame diversity in regulation as a bug, but many states now see it as a feature of a functioning democracy where they can tailor protections to their specific needs. It’s no longer about whether there will be a patchwork, but about which states will be the ones to set the initial, most influential stitches in that fabric. Lobbyists are finding that their old scripts just don’t resonate when legislators feel they have a moral mandate to act before the technology evolves beyond any hope of control.
There has been a visible shift in how tech lobbyists interact with statehouses, moving from a strategy of “freezing out” legislation to one of active engagement. What do you believe triggered this sudden realization that they could no longer afford to ignore state-level momentum?
It was a classic “David versus Goliath” moment that didn’t go the way the Goliaths expected, and the realization hit home when the tech industry’s massive financial advantage failed to translate into political compliance. At the beginning of 2026, there was this heavy, foreboding feeling in state capitals that a mountain of money—specifically the $100 million PAC backed by figures like Greg Brockman and major venture capitalists—would simply pounce on anyone who dared to regulate. However, when we saw primary results in places like Utah and New York, where candidates backed by those millions actually lost to pro-regulation incumbents, the aura of invincibility vanished. In Utah, a Republican representative handily defeated a well-funded challenger by leaning into his support for AI and kids’ safety, proving that data privacy and workforce protection are now winning campaign issues. Industry representatives suddenly realized that if they didn’t get on the train, they were going to get run over, leading to this frantic rush to “collaborate” on bills they were trying to kill just months ago. You can see the shift in their eyes during hearings; they’ve moved from dismissive posturing to a desperate attempt to streamline and harmonize rules before they lose all leverage.
Public perception of AI seems to have taken a sharp turn, with some legislators citing data that suggests 80 to 90 percent of constituents now view the technology as more bad than good. How is this “public angst” fundamentally changing the political math for lawmakers who might have previously leaned toward an industry-friendly, pro-innovation stance?
The political math has flipped because the “pro-innovation” label is no longer a safe harbor; it’s starting to look like a liability in the eyes of an increasingly skeptical electorate. When you have a bipartisan margin of voters—80 to 90 percent in some regions—expressing deep-seated anxiety about their jobs, their children’s mental health, and their personal data, a politician simply cannot afford to be seen as a puppet for Silicon Valley. I’ve spoken with lawmakers in California who mentioned that the polling they’re seeing is so lopsided it makes the tech lobby’s threats of running opposition candidates feel empty. It’s not just about the technology itself; it’s about the tangible impacts, like the surge in opposition to the massive data centers powering these models, which people see as a drain on local resources. This “angst” acts as a shield for legislators, allowing them to ignore the “Davids” of the tech lobby because they know their voters are behind them with an intensity we haven’t seen since the early days of the environmental movement. If you’re fundamentally out of sync with the sense of society, no amount of venture capital money can buy back that trust at the ballot box.
We are seeing a new dynamic where some tech companies, like Anthropic, are actually supporting certain regulations while smaller tech groups feel “outgunned.” How does this internal industry rift complicate the lobbying landscape, and who is actually winning the argument in the statehouses?
The internal rift has completely shattered the “industry vs. the world” narrative, creating a much more complex and volatile lobbying landscape where “Little Tech” feels betrayed by the giants. You have a fascinating situation where companies like Anthropic are advocating for strict regulations on models that require massive computing power and generate over $500 million a year, which their critics claim is a way to pull the ladder up behind them. This “regulatory capture” strategy, as some call it, effectively shields established players from smaller startups that can’t afford the same level of compliance or third-party auditing. Meanwhile, groups representing venture-backed “little tech” are crying foul, claiming they are now the “Davids” fighting against a new alliance of big developers and regulation advocates. This fragmentation is a gift to legislators because it allows them to pick and choose which industry voices to listen to, often siding with the “Big Regulation” camp because it aligns with public safety goals. The result is a legislative environment where the most powerful argument is no longer “don’t regulate,” but “regulate in a way that my specific business model can survive while my competitor’s cannot.”
Washington has historically been viewed as the ultimate authority on tech policy, but recent attempts to preempt state laws—like the “One Big Beautiful Bill Act”—have largely failed. Why has the federal government’s attempt to act as a “backstop” against state rules become what some call a “paper tiger”?
The federal government has lost its teeth in this fight because of a combination of gridlock and a series of high-profile strategic blunders that have emboldened state leaders. Last year, there was a real attempt to slide a 10-year moratorium on state AI regulation into a major defense bill, but it was stripped out at the eleventh hour because the backlash from governors and attorneys general was just too fierce. Then you have the executive orders from the White House, which initially sent a chill through statehouses with threats to sue or cut off funding, but as months passed in 2026 without a single lawsuit filed, states realized it was mostly posturing. Lawmakers in Vermont and Maryland, who were once worried about federal retaliation, are now moving forward with bills to rein in high-risk AI systems because they no longer see a coherent federal strategy. Even the “AI czar” positions have seen turnover and chaos, leaving a vacuum that state senators and assembly members are more than happy to fill. When Washington can’t even agree on basic data privacy standards, states feel they have not only the right but the obligation to step in and protect their citizens from the immediate risks of chatbot companions and algorithmic bias.
Governors have been described as the final line of defense against “onerous” state AI laws, with some vetoing bills or pushing for major amendments. Do you see this as a sustainable long-term strategy for the tech industry, or is the executive “backstop” also beginning to crumble?
The executive backstop is definitely showing cracks, and I don’t believe it’s a sustainable long-term strategy for Silicon Valley to rely solely on gubernatorial vetoes. While we saw governors in Virginia and California step in to block or amend safety bills recently, the political cost of doing so is rising every day as public opposition to the AI industry mounts. Even governors who were traditionally very tech-friendly, like those in Pennsylvania and Texas, are starting to feel the heat and are moving to curb things like new data center projects in response to local outcry. It’s one thing to veto a bill when it’s a niche technical issue, but it’s quite another when it’s a high-profile “kids’ safety” or “worker protection” bill that has 90 percent public support. If the “temperature” across the entire country has shifted, a governor who repeatedly kills popular safety legislation risks being painted as an industry shill, which is a dangerous label to carry into an election. We’re reaching a point where even the most pro-business governors will find it politically unwise to keep pulling the industry’s chestnuts out of the fire.
What is your forecast for the future of state-led AI regulation?
I anticipate that by the end of 2026, we will see at least a dozen states move beyond mere proposals to implement comprehensive, high-stakes AI safety frameworks that include mandatory third-party audits and strict liability for algorithmic harms. The “reverse federalism” strategy, where companies like OpenAI try to get states to coalesce around a single, weaker standard, will likely fail as states like Massachusetts and New York compete to pass the “toughest” laws in the nation. We are going to see a shift from broad, sweeping bills to very specific, sector-based regulations targeting things like psychological harms from AI chatbots in schools and the energy consumption of massive model training. The industry’s efforts to block these moves will continue to pivot toward “harmonization,” but they will be doing so from a position of weakness, effectively begging for a uniform floor rather than a ceiling on what states can do. Ultimately, the next two years will be defined by a “race to the top” in regulatory standards, where the states that move first will effectively dictate the national—and perhaps global—compliance environment for the next decade.
