Who Controls Technology?
The Kill Switch: How Governments Learned to Turn Off the Internet
In 2011, the Egyptian government switched off the internet for an entire country in under an hour — and the playbook they used has since been copied by dozens of states.
The Idea
Most people think of the internet as a decentralised, ungovernable web — the thing that was supposedly designed to route around damage, including political damage. That reputation is not entirely wrong, but it flatters the technology. The internet's physical and legal architecture contains natural chokepoints, and states have become expert at finding them. The key insight is that 'the internet' in any given country is not a wilderness of infinite paths. It flows through a surprisingly small number of licensed internet service providers, who are legally registered entities subject to national law. A government that controls those licences controls the pipe. Combine that with Border Gateway Protocol — the unglamorous routing system that tells data packets which direction to travel globally — and you have a system where a handful of authoritative commands can make an entire country's traffic vanish from the global network. This is not theoretical. Researchers at the network intelligence firm Oracle tracked 75 internet shutdowns in a single year across 25 countries. The methods vary: a full blackout, a social media throttle during elections, a targeted slowdown of encrypted traffic. Each technique requires a different level of technical sophistication, but all of them depend on the same underlying fact — that the internet, for all its apparent freedom, runs on licensed infrastructure that governments can compel. What makes this worth sitting with is that the same legal and technical mechanisms that enable shutdowns in authoritarian states are present, in softer form, everywhere.
In the World
On 27 January 2011, as protests gathered momentum in Cairo's Tahrir Square, Egypt's four main internet service providers — Telecom Egypt, Vodafone, Link Egypt, and Etisalat — simultaneously withdrew their routes from the global Border Gateway Protocol tables. To the outside world, Egypt simply stopped existing on the internet. The disconnection was not a hack or a technical failure; it was a series of phone calls. What followed revealed both the power and the limits of a shutdown. Mobile data and fixed-line internet went dark for five days. And yet the protests grew. Egyptians used landline dial-up numbers published by activists abroad, fax machines, and ham radio. A small team at Google and Twitter hastily built a 'speak-to-tweet' service — dial a number, leave a voicemail, it posts as a tweet — so that people without internet access could still reach the world. The shutdown accelerated international attention rather than suppressing it. President Mubarak resigned eighteen days later. The shutdown had failed its purpose — but that did not discourage other governments. Myanmar used a near-identical technique in the coup of February 2021, and went further by keeping targeted regions blacked out for over a year. Iran, during the 2019 protests over fuel prices, executed a shutdown so technically refined that it kept government and banking infrastructure running internally while severing the country from the global internet — a more surgical and, in the short term, more effective approach.
Why It Matters
This is not a story about faraway authoritarianism. It is a story about infrastructure and leverage — and infrastructure is everywhere. Every time a government grants a licence to a telecoms company, negotiates a data-localisation law, or requires that encryption keys be held domestically, it is quietly accumulating the same kind of leverage that Egyptian officials exercised in 2011. The mechanisms are present in democracies too; the question is what guardrails, legal traditions, and political costs constrain their use. For anyone who builds things online, works in communications, or simply relies on the network for their livelihood or safety, understanding that the internet is a regulated utility — not a force of nature — changes how you think about resilience, about jurisdiction, and about what 'free and open' actually means in practice. It also sharpens a harder question: in the places where shutdowns are used most aggressively, the populations most harmed are often those with the least power to challenge them. The technology did not create that inequality, but it has given it a new and very efficient shape.
A Question to Ponder
If the infrastructure of communication can be switched off by the same legal mechanisms that keep it running, what does it actually mean to call the internet a public resource?
Get a new one of these every morning.
Start learning with Thinkable