1960s: An idea born of Cold War anxiety
The internet begins not with a business plan but with a fear: that a single attack could destroy America's centralised communications. Researchers funded by the US Department of Defense proposed a radical alternative — a network with no centre, where messages would be broken into small packets, each finding its own route to its destination, and reassembled on arrival. If one path failed, another would do.
The key insight, called packet switching, was developed independently by researchers in the United States and Britain. In 1969, the first link of the resulting experimental network, ARPANET, connected computers at four American universities. The first message famously crashed the system partway through — the network transmitted "LO" before failing, intending "LOGIN". It was an inauspicious start for the most consequential communication system in history.
1970s–1980s: The plumbing gets standardised
Through the 1970s, ARPANET grew and inspired rival networks, but they could not talk to each other. The breakthrough was a common language: TCP/IP, a set of rules for how packets travel and how networks interconnect, developed by Vint Cerf and Bob Kahn. On 1 January 1983, ARPANET switched to TCP/IP permanently — a date some historians call the internet's true birthday.
Electronic mail, invented in the early 1970s, became the network's first killer application. By the 1980s, use had spread from military contractors to universities worldwide. But the internet remained a text-based tool for specialists. Using it required technical knowledge, and nothing about it suggested a mass medium.
1989–1991: The Web is invented
The turning point came from an unlikely place: a physics laboratory in Switzerland. Tim Berners-Lee, a researcher at CERN, proposed a system of linked documents that could be navigated by clicking — the World Wide Web. He built the first web browser, the first web server, and the first website, which went live in 1991.
Crucially, CERN released the web's underlying technology royalty-free. Anyone could build on it without permission or payment. That single decision — openness over ownership — shaped everything that followed. The internet is the infrastructure; the web is the system of pages and links most people mean when they say "the internet".
1990s: The web goes public
The web needed an easy on-ramp, and in 1993 it got one: Mosaic, the first widely used graphical browser, which displayed images alongside text and could be installed by non-specialists. Usage exploded. Netscape, founded by Mosaic's creators, commercialised the browser; Microsoft answered with Internet Explorer, igniting the first browser war.
Businesses rushed online. Amazon (founded 1994) and eBay (1995) pioneered online commerce; Google (1998) set out to organise the web's chaos with a better search engine. By decade's end, investors were pouring money into any company with a website, inflating the dot-com bubble. When it burst in 2000, hundreds of firms collapsed — but the infrastructure, the users and the useful survivors remained.
2000s: Broadband, social and mobile stirrings
The 2000s made the internet fast and social. Broadband replaced dial-up in many households, making video and rich media practical. A second generation of companies built the participatory web: Wikipedia (2001) proved strangers could build an encyclopaedia together; Facebook (2004) and YouTube (2005) gave everyone a publisher's tools; Twitter (2006) compressed public conversation into short bursts.
The decade's most consequential launch came in 2007, when the iPhone put a capable web browser in people's pockets. Mobile internet had existed before, but the modern smartphone made it effortless — and shifted the internet's centre of gravity from the desk to everywhere.
As smartphones multiplied, computing moved into the cloud — vast data centres rented by the minute — which let tiny startups serve millions of users. Streaming replaced ownership for music and much of video. A handful of platforms grew to extraordinary scale, raising new questions about privacy, competition and the power of private companies over public conversation.
The decade also saw the internet truly go global. Billions of people in Asia, Africa and Latin America came online, many experiencing the internet first through a phone rather than a computer. The web's centre of cultural gravity shifted decisively away from its American and European origins.
2020s: The AI era
The current chapter is being written around artificial intelligence. Powerful AI models, trained on vast amounts of data and served from the cloud, now write, code, translate and create images on demand — capabilities that would have seemed like science fiction a decade ago. The technology raises profound questions about work, creativity, truth and trust online.
Looked at whole, the internet's history is a story of decentralised beginnings, commercial concentration, and now a new wave of disruption. From "LO" in 1969 to AI assistants in billions of pockets, the arc has been faster than anyone predicted — and it is not finished.
The invisible machinery
Behind the eras and the household names sits machinery most users never think about. The domain name system quietly translates memorable addresses into numeric ones, like a phone book for the entire planet. Undersea cables — hundreds of them, carrying the overwhelming share of intercontinental traffic — are the internet's true backbone, more important than any satellite. Content delivery networks copy popular data to servers near you, so a video loads quickly whether you are in London or Lagos.
None of this was inevitable. Each layer was built by people solving the problem in front of them, and each became infrastructure for the next wave. The internet feels like a single thing, but it is really a stack of agreements — technical, commercial and social — renewed daily by millions of participants who will never meet.
What the timeline teaches us
A few patterns repeat across the decades. First, openness wins: the decisions to make TCP/IP and the web royalty-free created more value than any proprietary alternative could have captured. Second, the killer application is never obvious in advance — email, the browser, search, social networks and AI assistants each surprised the previous generation's experts. Third, every wave of decentralisation seems to be followed by concentration, as new gatekeepers replace old ones.
Perhaps the most useful lesson is humility about prediction. In 1990, almost nobody foresaw the web; in 2005, almost nobody foresaw the smartphone's dominance; in 2015, few foresaw the current AI wave. The internet rewards those who build for what is possible rather than what is familiar — and it has a habit of making confident forecasts look quaint within a decade.
