AI is new. Our reaction to it is not.

Core argument: AI deserves serious attention to real risks, but panic is a poor operating system. Across electricity, automobiles, telephones, broadcast media, and the Internet, people repeatedly worried about physical danger, nervous exhaustion, moral decline, social fragmentation, and elite control. The practical lesson is to distinguish real risks from generalized fear, and to build transparency, competition, accountability, and widespread literacy around powerful new tools.

Every generation meets a transformative technology twice: first as a threat, then as infrastructure. The first encounter is emotional. The new tool feels unnatural, too fast, too powerful, too invasive. The second encounter is practical. Society learns where the risks are real, where the fears were exaggerated, what practices work, and what habits must change.

Artificial intelligence feels unprecedented because it performs tasks long associated with human language, prediction, creativity, and expertise. That makes it psychologically unsettling. But our reaction is not entirely new. Previous generations worried that new communication systems would overload attention and erode relationships; historians note that nineteenth-century anxieties about speed, stress, and mediated life closely resemble today's concerns.¹

The lesson of history is not that we should ignore AI risks. Electricity, automobiles, broadcast media, and the Internet each carried real consequences.^{3 5 9 10 14} The better lesson is that every major technology produces a mixture of justified concern, exaggerated panic, commercial interests, and eventual adaptation. A mature society learns where to focus its attention.

The Pattern: Panic, Proof, Rules, Normalization

Across technological history, similar anxieties recur. People worry the new tool will harm bodies, overstimulate minds, corrupt children, destroy jobs, weaken community, empower elites, or obscure truth. These fears are often amplified by speed of adoption, media coverage, and interested parties.

The historical pattern is rarely outright rejection. It is a cycle of enchantment and dread, visible problems, adaptation through better practices, and eventual normalization. This does not guarantee good outcomes, but it shows adaptation is a social process. For AI, the central question is not "Should we be afraid?" but "Which risks are real, who bears them, and how can transparency help manage them while preserving the benefits?"

Electricity: Invisible Power and the Fear of Nervous Exhaustion

Electricity combined wonder with fear. It was invisible and poorly understood, making it easy to imagine as both miraculous and dangerous. Physician George M. Beard linked modern "nervous exhaustion" to accelerated life, including new technologies.² Early electrical systems were indeed hazardous. Yet adoption surged as the technology improved: from 8% of U.S. homes in 1907 to 68% by the late 1920s, and over 90% of farms by 1953.^{3 4}

The real progress came from engineers, insurers, utilities, and market incentives creating safer systems: wiring standards, circuit breakers, professional practices, and better infrastructure. AI needs a similar shift from mystery to manageable tool - not through top-down control, but through transparency so people understand when AI is being used, how it works, where it fails, and who stands behind it.

The Automobile: Speed, Status, and the Politics of Safety

U.S. motor vehicle registrations exploded from about 8,000 in 1900 to 26.7 million in 1930.⁵ Early reactions included medical warnings about "Dementia Automobilis" and criticism that cars symbolized elite arrogance.^{6 7}

Society adapted through a combination of market forces, insurance, consumer demand, and targeted rules: licensing, traffic systems, safety standards, and liability. Competition and innovation - seatbelts, airbags, better vehicle design - dramatically improved safety over time. AI is in a similar early phase. Benefits are exciting but uneven. The pragmatic response is not to slow the road, but to promote transparency in high-stakes uses, protect against clear harms, and let competition drive better tools.

The Telephone: Intimacy, Interruption, and Mediated Life

Early users feared overhearing, deafness, and constant availability. Social norms and etiquette eventually evolved - caller ID, voicemail, calling hours, and other practices - without banning the technology.⁸

AI raises a sharper version of this: it can blur human and automated communication. The right approach is not blanket regulation, but clear disclosure in contexts where trust and identity matter, combined with competition that rewards reliable, transparent systems.

Radio and Television: Mass Media, Children, and Attention

Radio and television created mass attention economies and sparked familiar fears about children, addiction, and passive minds.^{11 13} Many concerns were overstated, including the "War of the Worlds" panic.¹² Yet society adapted through cultural norms, media literacy, competition, and voluntary standards rather than outright suppression.

AI amplifies these dynamics with generative and personalized content. The answer lies in provenance tools, transparency, age-appropriate design, and vigorous competition, not centralized control of the feed.

The Internet: Isolation Panic, Connection, and Evidence

Internet adoption went from 52% of U.S. adults in 2000 to 96% in 2025.¹⁴ Early "Internet Paradox" fears largely faded as people adapted.¹⁵ The lesson: broad claims often give way to more precise understanding. Access plus literacy proved more powerful than fear.

AI will face its own digital divide - not just who can use it, but who can use it wisely. Widespread AI literacy, verification habits, and competitive markets that provide broad access and reward trustworthy systems will help bridge this gap.

AI: The Old Psychology Meets a Genuinely Powerful Tool

AI adoption is growing rapidly. Pew found 34% of U.S. adults had used ChatGPT by mid-2025.¹⁶ McKinsey reported 88% of organizations using AI in at least one function.¹⁷

Frameworks like NIST's AI Risk Management Framework and the OECD AI Principles offer voluntary, risk-based guidance focused on transparency, robustness, and accountability.^{18 19 20} These are useful starting points precisely because they emphasize evidence over blanket rules.

Conclusion: Do Not Fear the Future - Build with Clarity and Competition

History shows fear is a poor operating system. Panic flattens important distinctions. Pragmatism separates real harms from symbolic ones. It asks: Where is transparency needed most? How can competition drive better outcomes? How do we equip individuals to participate effectively?

AI combines elements of past technologies: the invisible power of electricity, the speed of the automobile, the reach of broadcast media, and the networked nature of the Internet. The wise response is not alarm or surrender, but deliberate, light-touch adaptation: clear disclosure where it matters, strong competition to reward trustworthy systems, broad AI literacy so citizens can leverage the technology confidently, and institutions that learn from real-world results rather than preempt them.

The future will be shaped by treating AI as what every major technology becomes: a human tool, made better through transparency, innovation, and informed use.