In recent decades, we have created a new zone of liberty: the digital world. Indeed, it might be said that the internet was conceived in liberty, free of the regulation that has bound and hampered the industries of the physical world, industries ranging from finance to healthcare to energy to transportation; even some legacy communications industries, such as broadcasters, work within the confines of a complex regulatory code. In a certain sense, the light-touch regulatory regime which has thus far obtained in America echoes the benign neglect practiced by the British Empire toward its American colonies in the 17th and 18th centuries. Left alone, American innovators have thrived, evidenced by the fact that the six largest companies in the world by market cap all belong to the technology sector, and all are American.1https://companiesmarketcap.com/
The period of benign neglect seems likely to be slipping away, however, as lawmakers in Washington, D.C., and in state capitals fret about the social effects of new technologies. Social media and artificial intelligence (AI) have become two special objects of disapprobation. The worries that attach to each technology vary, but both implicate the same cornerstone of American liberty: free speech.
Digital technologies have made it possible, for the first time, “to organize the world’s information and make it universally accessible and useful.”2https://www.businessinsider.com/laszlo-bock-on-google-mission-statement-2015-4?utm_source=chatgpt.com The compact rectangles on our desks and in our pockets allow us to learn almost anything about anything, and to converse and debate with others across the country and across the world — and to do so freely, without the interference of the state.
Recent proposals for social media and AI threaten to constrain, to close, and to suffocate the free and open internet that has thus far flourished in America. Of particular concern are recent proposals in the states, which, for better and for worse, have not succumbed to the torpor that afflicts Congress; to the contrary, the states have proven themselves hyperactive. For instance, in 2025 alone, state legislatures entertained more than 1,000 proposals to regulate AI.3https://www.lawfaremedia.org/article/1-000-ai-bills–time-for-congress-to-get-serious-about-preemption “In the 2025 legislative session, all 50 states, Puerto Rico, the Virgin Islands, and Washington, D.C., have introduced legislation on this topic this year,” National Conference of State Legislatures noted in July.4https://www.ncsl.org/technology-and-communication/artificial-intelligence-2025-legislation “Thirty-eight states adopted or enacted around 100 measures this year.”
Regrettably, in too many cases, lawmakers seem to understand neither the technology they hope to regulate nor the far-reaching second- and third-order effects of their proposals. Indeed, they very often seem unaware that the freedom of Americans to think, learn, and speak freely is in danger, likely to be stifled by well-meaning, though misguided, bids to control the dissemination of information online.
A better conception of the issues at hand must be discovered and made manifest in policy.
Human beings do not perceive information, form opinions, or converse in isolation. From the beginning, children’s consciousnesses are contoured by the upbringing they receive from their families. As we grow, our thoughts and beliefs develop in conversation, as it were, with our friends, our teachers, our communities. Our investigations of the world and our quests for knowledge and understanding are mediated by the ideas we are taught, the books we read, the media we consume, the little platoons in which we live, and an array of other inputs and institutions. In short, human understanding is inexorably enmeshed in an astoundingly rich and complicated context. The fact that the pursuit of knowledge and truth is a collaborative and communal endeavor cannot be avoided, and it is not to be regretted.
Social media and AI tools are just two — albeit novel — species of mediating institutions which inform human understanding and the creation and conveyance of information. Indeed, the intellectual life of humanity has always carried on within the context and constraints of such mediators. Before X and Facebook, the news was selected, filtered, and molded by legacy newspapers and television reporters; before Google and ChatGPT, the products of research were limited by the interpretations of encyclopedia editors; before Americans made themselves sick over cycles of online outrage and hysteria and “misinformation,” worrywarts vituperated the “confusing and harmful”5https://slate.com/technology/2010/02/a-history-of-media-technology-scares-from-the-printing-press-to-facebook.html wash of information unleashed by the printing press and the “dangerous and injurious” effects of reading novels;6https://x.com/PessimistsArc/status/1804163304480264271 before social media was supposed7https://lawliberty.org/the-dubious-case-for-regulating-internet-habits/ to have corrupted the youth, the pager was the Socrates of the day;8https://newsletter.pessimistsarchive.org/p/the-forgotten-war-on-beepers and before online echo chambers fomented partisan outrage and superstition, insular small-town life produced the same effects.
Online platforms differ from their predecessors in that, for all the distortion of information and understanding they are said to cause, they allow individuals to venture beyond narrow ideological siloes and to discover something different. If Twitter is not to your liking, Bluesky awaits. If ChatGPT seems inordinately sanitized for your taste, Grok will supply quite another set of answers to your queries. Put differently, the phenomenon of bubbles and bad information, of confirmation bias and partisanship, is a very old thing, endemic to the human condition. The chance to escape, to get more information, to weigh competing claims and uncover other facts is one of the revolutionary characteristics of the internet age. It is to be celebrated, not destroyed.
A question remains: Who is to decide? Will the government determine what sorts of speech are fit to be put before the American people,9https://www.nationalreview.com/2024/09/why-conservatives-are-wrong-to-support-the-kids-online-safety-act/ managing the content moderation of social media platforms10https://www.protectingtaxpayers.org/free-speech/content-moderation-at-the-supreme-court-free-speech-on-and-for-social-media-platforms/ and the outputs of AI?11https://www.nationalreview.com/2025/07/dont-teach-the-robots-to-lie/ Will it condition Americans’ access to online platforms on a show-your-papers regime,12https://www.realclearmarkets.com/articles/2025/05/21/online_age_verification_is_the_show_your_papers_of_the_digital_age_1111457.html placing a policeman at the door of every social media platform13https://www.realclearmarkets.com/articles/2025/07/10/show_your_papers_threatens_life_on_the_internet_1121444.html and AI chatbot14https://x.com/Protectaxpayers/status/1967965112230392133 and eviscerating the privacy and data security of American adults, children, and families?15https://www.rstreet.org/commentary/the-fundamental-problems-with-social-media-age-verification-legislation/
Or will private businesses, subject to market forces and the values of their users, be left free to innovate and experiment, to create digital products that promote the good of individuals and of the nation? This, it must be admitted, requires a certain degree of trust; but that trust is not unfounded. Just this year, dissatisfaction with Facebook’s too-tight content moderation prompted Meta to reform the platform’s policies.16https://www.protectingtaxpayers.org/technology/taxpayer-watchdog-reacts-to-metas-new-content-moderation-policies-2/ In 2022, driven by the same dissatisfaction among Americans, Elon Musk acquired Twitter and subsequently renamed it X. Those who disliked X’s new sensibilities flooded to Bluesky, and those who had become tired of politics-dominated social media altogether found refuge in Meta’s Threads. The journey toward a more perfect digital information ecosystem has proved fraught, circuitous, and iterative, but it seems far better suited to yield something approaching “good” than the blunt-force, unresponsive mechanism of legislation and central planning.
America has hitherto believed that disputes about the truth must, in the main, be settled by debate and civil society, not the prescriptions of the state. “Conscience is the most sacred of all property,” James Madison wrote.17https://press-pubs.uchicago.edu/founders/documents/v1ch16s23.html The right to a free conscience must be accompanied by what John Stuart Mill termed its cognate right to free speech. That faith — that heritage — ought not to be abandoned now.
“You do not defend a world that is already lost,” wrote Garet Garrett in 1938, after the shadow of the New Deal had fallen across the land. “When was it lost? That you cannot say precisely. It is a point for the revolutionary historian to ponder. We know only that it was surrendered peacefully, without a struggle, almost unawares.”
The free and open internet is not yet lost, but one senses that it might soon be. With each session undertaken by state legislatures, it seems ever likelier that a patchwork regulatory morass will dull the vital innovative force of the technology sector and fetter the constitutional — and, more importantly, the natural — rights of Americans. Digital technologies transcend borders, and the effects of their regulation often cannot be contained within the state or jurisdiction in which a regulation originates.18https://www.lawfaremedia.org/article/1-000-ai-bills–time-for-congress-to-get-serious-about-preemption In far too many cases, this is being done, indeed, “without a struggle, almost unawares,” with little regard for the freedom which is being dismantled.
A better understanding of the purpose of digital platforms, a more careful reading of history, and close attention to the unintended — though not unwarned-of — consequences of a regulatory revolution underway can conserve the freedom of the digital world.
David B. McGarry is the research director at the Taxpayers Protection Alliance.
