1969: Apollo 11 lands on the Moon

On 20 July 1969, the lunar module Eagle touched down in the Sea of Tranquility, carrying Neil Armstrong and Buzz Aldrin. Behind the iconic “one small step” stood a compact, mission-critical computer: the Apollo Guidance Computer (AGC). Designed at MIT’s Instrumentation Laboratory under Charles Stark Draper, the AGC weighed about 32 kilograms, ran at roughly 1 MHz, and held just 36 kilowords of core-rope memory—yet it guided the spacecraft from Earth orbit to the lunar surface and back.

The AGC was not “intelligent” in the modern sense, but it embodied ideas that later fed directly into artificial intelligence. It executed real-time priority scheduling so that the most urgent tasks—landing radar, attitude control—never lost the CPU. When the famous 1201 and 1202 alarms flashed during descent, the software calmly shed lower-priority jobs and kept the landing sequence alive. That resilient, interruptable architecture prefigured the interrupt-driven and multi-agent systems common in today’s robots and autonomous vehicles.

Software engineer Margaret Hamilton led the team that coined the term “software engineering” while building the AGC code. Their insistence on exhaustive simulation, asynchronous executive design, and human-rated reliability became foundational practices. These same disciplines now govern safety-critical AI: verifiable planners, runtime monitors, and fail-operational behaviour in self-driving cars, surgical robots, and spacecraft autonomy stacks.

The intellectual thread runs further back through Norbert Wiener’s cybernetics and the wartime fire-control computers that inspired both Wiener and the AGC designers. Feedback, state estimation, and closed-loop control—once the province of analog servomechanisms—were digitalized aboard Apollo and later generalized into the estimation and control loops inside modern learning systems.

Today every Mars rover, every drone that lands itself, and every large language model deployed in a real-time loop inherits lessons first proven under the harsh constraints of 20 July 1969. The Moon landing remains a landmark demonstration that carefully engineered computation can extend human reach farther than unaided intuition alone—an enduring bridge between classical control, reliable software, and the adaptive intelligence we now call AI.