The relationship between consumers and artificial intelligence has entered a strange new phase. Public opinion surveys consistently record deep skepticism, outright hostility, and a growing sense that AI is being forced upon an unwilling population. Yet the same people who declare they hate AI continue to use it — to generate text, summarize documents, edit photos, and automate tasks. This paradox sits at the center of the current technology landscape: widespread adoption coexists with widespread resentment. The explanation for this contradiction is not that consumers are hypocrites. The real story is more nuanced and far more revealing about the strategies of the companies pushing AI into every corner of daily life.
The Real Target of Public Anger: Not AI but the Corporate Agenda Behind It
The visceral reaction many people express when they say they hate AI is rarely aimed at the technology itself. What fuels the anger is the relentless drive of the companies that develop and deploy AI to embed it into as many products, services, and workflows as possible, often without meaningful consent or genuine user benefit. These same companies have been telling the public to brace for the biggest social and economic upheaval in generations — a message that is inherently unsettling. To understand the depth of the backlash, it helps to ask a simple question: how else are people supposed to feel when trillion-dollar corporations simultaneously warn that AI will transform everything and then proceed to cram it down every available channel?
The lesson of the social media era is that once a technology becomes essential, the public learns to live with its flaws.
The framing from tech leaders has often oscillated between utopian promises and apocalyptic warnings, leaving little room for a measured public conversation. Consumers are not stupid. They sense that the enthusiasm for AI from the supply side is not primarily about improving their lives — it is about extracting value, capturing markets, and building moats. The technology itself, whether a large language model or a recommendation algorithm, is neutral. The business models and deployment strategies are not. That distinction is critical for anyone trying to navigate the current moment.
History Repeats: The Social Media Techlash That Did Not Stop Adoption
This dynamic is not unprecedented. The pattern played out in near-identical fashion during the rise of social media. Billions of people flocked to Facebook, Twitter, and Instagram even as a mounting techlash targeted the companies behind those platforms. The same was true for Google Search — universally used yet widely distrusted. The reason was not that users were indifferent to the harms. It was that the switching costs were prohibitive. Quitting a social network meant abandoning years of content, personal connections, and an entire digital identity. The choice was to either tolerate the platform’s behavior or start over from zero. Most chose the former.
AI adoption today faces a similar structural constraint, but with an important difference. In the social media era, there was hardly any regulatory or market pressure to change how platforms operated. Laws were slow, antitrust enforcement was weak, and alternatives were limited. For AI, the window of opportunity to shape outcomes is still open. The regulatory landscape is already shifting, and the availability of credible open-source alternatives means that consumer choice — and the market pressure it generates — could play a far more significant role than it ever did in the social media story.
A Legislative Avalanche: More Than 2,100 AI Bills Across 50 US States
The political appetite for reining in AI is dramatically higher than it ever was for social media at a comparable stage of adoption. All fifty US states have either passed or proposed laws governing the development and deployment of AI, creating a patchwork of more than 2,100 bills across the country. That figure represents a tenfold increase in just three years. The sheer volume of legislative activity signals that lawmakers — and, by extension, voters — are paying close attention and are unwilling to leave the trajectory of AI solely in the hands of the companies building it.
This regulatory flurry is not uniform. Some states are focusing on transparency requirements, others on algorithmic accountability, and still others on consumer protections around automated decision-making. The result is a fragmented but rapidly evolving legal environment that imposes real costs and constraints on AI companies operating across jurisdictions. Whether this patchwork will coalesce into a coherent national framework remains to be seen, but the direction is clear: the era of unchecked AI deployment is drawing to a close. The question is what comes next.
Why Regulatory Pressure Alone Is Not Enough — And Where Open Source Changes the Equation
Regulation alone will not solve the adoption-dislike paradox. Lawmakers can mandate disclosures and set guardrails, but they cannot force consumers to feel good about a technology they did not ask for. The deeper issue is one of trust and agency. People want to feel that they have control over the tools they use and that those tools are designed with their interests in mind, not just the bottom line of a corporate parent.
This is where the emergence of high-quality open-source AI models becomes a wild card. Unlike the social media era, when alternatives to Facebook or Twitter were either inferior or nonexistent, the AI market today already boasts a number of top-tier open-source alternatives to the models offered by Google, OpenAI, and Anthropic. These open-source models are not just academic curiosities — they are competitive in performance, available for free, and can be run locally or on private infrastructure. For a consumer or a business that wants to use AI without feeding data into a corporate black box or being locked into a proprietary ecosystem, open source offers a genuine alternative.
The market pressure created by these alternatives is real. If a significant number of users begin to migrate to open-source models because they offer more transparency, lower cost, or greater control, the dominant players will have to respond — either by changing their own behavior or by losing share. That dynamic never materialized for social media because network effects made it nearly impossible to leave. For AI, the switching costs are lower. You can start using a different model tomorrow without losing your data or your connections. That changes the power balance.
The CEO Who Doesn’t Like AI — And What He Said About the Path Forward
One of the most telling perspectives on this moment comes from the CEO of a firm called Springboards, a company that is actively developing a new large language model despite its leader’s personal ambivalence toward the technology. When asked why his company was pursuing the project if he did not really like AI, the CEO offered a response that cuts to the heart of the matter. He acknowledged that there is no walking back from large language models — the technology has been unleashed and can be un-invented. But, he added, you can still make them do something different.
That is a crucial insight. The irreversibility of LLMs does not mean the outcome is predetermined. The technology is a platform, not a destiny. The companies that build and deploy these models have choices about what they emphasize — accuracy over hype, transparency over opacity, user agency over corporate control. The CEO’s comment reflects a pragmatic realism: since the technology is here to stay, the responsible path is to steer it toward constructive ends rather than to rail against its existence.
What Consumers Actually Want: Clear Limits and Honest Positioning
If the Springboards CEO’s perspective resonates, it is because it aligns with what many users are implicitly asking for. The public does not need AI to promise to take over the world. It needs AI that is clear about what it can and cannot do. It needs tools that do not posture as omniscient or omnipotent but instead present themselves as limited, fallible, and designed to serve specific purposes. The grandiosity of the rhetoric around AI — the talk of artificial general intelligence, of sentience, of imminent radical transformation — is itself a major driver of the dislike that people express.
When a chatbot claims to be conscious or a marketing deck suggests that AI will replace half the workforce, the public responds with fear and resentment. Those reactions are not irrational. They are a direct response to the way the technology is being sold. A more restrained, honest, and capability-centered approach would likely reduce the backlash significantly, even if adoption continues to grow. The goal should not be to make everyone love AI. It should be to make AI boring, reliable, and useful — the same trajectory that electricity, the internet, and the smartphone eventually followed.
No Inevitability: The Power of Choice and the Path to a Different AI
The core misconception that the industry has cultivated is that the current trajectory of AI is inevitable — that the technology is on a one-way march toward ever greater power, ever deeper integration, and ever less human control. That is a self-serving narrative. Trillion-dollar companies are hard to move, but they are not immovable. Regulatory pressure, consumer choice, and the availability of open-source alternatives all create leverage points that can alter the direction of development.
The next few years will determine whether AI evolves into a tool that serves human agency or one that erodes it. The deep dislike that so many people feel is not a sign that they reject the technology wholesale. It is a signal that they want a different relationship with it — one built on transparency, consent, and genuine utility rather than on corporate diktat and hype. The companies that listen to that signal will be the ones that survive the backlash. Those that ignore it will find that the adoption curve flattens, not because people stop using AI, but because they find alternatives that respect what they actually want.
The lesson of the social media era is that once a technology becomes essential, the public learns to live with its flaws. The lesson of the AI era may be that the public does not have to accept those flaws at all. The market is still fluid. The laws are still being written. The technology is still young. What comes next is not as inevitable as the companies often imply. The hope is that we see a lot more of something different — AI that is honest about its limits, designed for human benefit, and content to be a useful tool rather than an aspiring overlord.
