History of cybersecurity

Cybersecurity was born with open networks: the more systems connect, the more fragile implicit trust becomes. Its history tells the story of the gradual professionalisation of attack and defence.

The idea of self-replicating programs predates their existence: John von Neumann laid out the theory as early as the 1940s. In 1971, on ARPANET, the experimental Creeper program moved from machine to machine displaying a message; Reaper was written soon after to erase it: the first "virus", the first "antivirus".

The term computer virus was formalised in 1983 by the researcher Fred Cohen, the very year the film WarGames firmly planted the hacker in the collective imagination.

On 2 November 1988, Robert Tappan Morris released an experimental worm that escaped his control and paralysed about 6,000 machines, nearly 10% of the Internet at the time. The incident founded the discipline: creation of the first CERT at Carnegie Mellon and the first major conviction for computer fraud.

The initial lesson was set: a network designed among people who trust each other does not survive scaling up intact.

With the spread of the PC and the Internet, the threat changed scale: the birth of the antivirus industry, the high-profile arrest of Kevin Mitnick in 1995, then the ILOVEYOU worm in 2000, which infected tens of millions of machines through a simple attachment.

In the 2000s, hacking stopped being an insiders' sport: it became a business made of botnets, spam, banking theft, resale of access and an underground economy.

2007: Estonia suffered waves of denial-of-service attacks that paralysed banks and government services. 2009: France created ANSSI, its national cybersecurity agency, attached to the Prime Minister's office. 2010: discovery of Stuxnet, a worm attributed to the United States and Israel, designed to physically destroy Iranian nuclear centrifuges.

2013: Edward Snowden's revelations exposed the scale of mass surveillance and changed attitudes to encryption. Then came Sony Pictures (2014), Mirai (2016), and the dark year of 2017: WannaCry and NotPetya, which showed that a cyberattack can stop hospitals, factories and global supply chains.

The recent period has brought break after break: SolarWinds (2020), Colonial Pipeline (2021), Log4Shell (2021), Change Healthcare (2024), massive data leaks at France Travail and Free, hospitals held to ransom. In July 2024, the global CrowdStrike outage showed, without any attack at all, the systemic fragility of digital infrastructure.

At the end of 2025 came a new turning point: the revelation of cyber-espionage campaigns largely orchestrated by AI opened the era of fast, parallel and partly autonomous attacks.