The near disasters before AI

The people building AI now warn it could kill everyone by the end of the decade. A researcher who quit the AI company Anthropic said the people doing the work genuinely believe that. Ask them how, exactly, that would happen, and the answers tend to land on three things: a bioweapon, a poisoned water supply, or a mistaken missile launch.

Each of those has already come close — without any AI. What stopped those near misses says more about the real danger of AI than any forecast does. In each one, the thing that failed was a machine, and the thing that saved everyone was a human paying attention.

Shortly after midnight on Sept. 26, 1983, a siren went off in a bunker south of Moscow, and a big red screen in front of Stanislav Petrov, the lieutenant colonel on duty, read "launch." He had minutes to decide whether the United States had just fired five nuclear missiles at his country. His job that night was to watch the Soviet Union's early-warning satellites and pass any launch straight to the military and political leadership. The system reported one American missile, then a second, a third, a fourth, and a fifth, and it rated the warning's reliability at the highest level, Petrov told the BBC years later.

Petrov didn't believe the computer, and he was right. Nothing had been launched.

Sponsored

Staying put feels easy, but it doesn’t always mean you’re getting a good rate. That’s why drivers have compared more than 19 million quotes on Compare.com, and customers rate it 4.6 stars. You’ll see real, personalized quotes from 120+ insurers side by side, and you could save up to $900 a year.* No spam, no phone calls after you sign up, and it’s completely free. Your wallet will thank you for the 2 minutes it takes to get a personalized quote.

Semafor Flagship is the essential email briefing that connects global developments across politics, economics, climate, AI, and more. Each edition provides context and insight on the ten most important stories happening around the world.

The team behind PHP, the programming language that runs roughly 70% of the world's websites, released patches last week for two critical flaws that let attackers run their own code on a web server without logging in, effectively taking it over. Organizations rushed to install the fixes, and some took websites offline in the meantime.

Every PHP installation that handles web traffic is at risk, the project said, and servers that run PHP inside the Apache web server face the greatest danger. PHP versions 8.5.12, 8.4.27, 8.3.36, and 8.2.35 close the two holes.

The Cybersecurity and Infrastructure Security Agency urged organizations to patch the flaws, which it calls "php2shell," starting with systems exposed to the internet. Organizations that can't update, or that run PHP 8.1 or older, should take their PHP-based systems offline until they're fixed, the agency said.

Some site owners pulled services offline, including parts of Wikipedia and 4chan. Government websites and services in the U.K., Finland, and Japan were also disrupted. Some sites may stay down for a while because fixes vary from one piece of software to the next and have to be tested before systems go back online.

Sponsored