For more than a decade, conservative activists held that a single atomic bomb could, without killing Americans directly, lay waste to the country — a claim that built fervor for war with Iran.
The test detonation in 1962 of a hydrogen bomb lit up the night sky over Honolulu and knocked out some of the city’s electrical systems. Experts linked the unanticipated damage to what they came to call the electromagnetic pulse.
This is no scoffing matter. The accelerating proliferation of advanced nuclear weapons is bringing ‘end times’ closer by the day
Humankind has fearfully and sometimes joyfully anticipated the end of the world ever since it began. The Bible’s Book of Revelation speaks of “end times”, a period of tribulation, war, famine, plague, floods and environmental catastrophe that may seem a less absurd prospect today than perhaps it once did. Islam, too, posits a fateful day of judgment or day of reckoning. Christian evangelicals believe that, in the final rapture, a chosen few – the righteous or the “elect” – will be saved while sinners burn. It’s a minority view. Matthew’s gospel predicts most people will defy, ignore or ridicule warnings of imminent Armageddon. Peter, another of Jesus’s disciples, agreed: “In the last days, scoffers will come, scoffing and following their own evil desires.” Yet such worldly scepticism does not mean a universal cataclysm is less likely. It may make it more probable.
Donald Trump’s blanket denials of the dangers ostensibly posed by artificial intelligence, whose uncontrolled development is seen by some as the latest harbinger of apocalypse, could be interpreted as fulfilling Peter’s prophecy. This US president, the nearest thing the world has to a one-man doomsday machine, is a great one for scoffing and evil desires. In Robert Harris’s 2019 novel, The Second Sleep, a worried scientist lists half a dozen ways the world might end. They include devastating climate change, pandemic disease, a giant asteroid collision with Earth, or some other overwhelming calamity. Few heed the fictional scientist’s fears. In the event, what destroys him and almost everybody else is a disturbingly plausible cyber crash – a global digital and internet meltdown.
Simon Tisdall is a Guardian foreign affairs commentator
Newly established commission identifies 17 catastrophic threats to human health but says ‘realistic route to act’ on each exists
Humans could become extinct by the year 2100 due to malicious use of AI or nuclear conflict, according to a new global risk assessment.
They top a list of 17 major threats to human health over the next 75 years, which also includes cuts to aid spending by wealthy countries, the climate crisis and childhood malnutrition.
Last week, Anthropic released a long and detailed document describing current misuses of their Claude models. I’m still reading it, but I wanted to flag this:
We identified a cell of threat actors based in northern Yemen running three weapons development programs: a guided rocket that used a commodity phone-class flight computer with final-phase homing guidance; a multi-stage ballistic missile with a stated range goal above 2,000 km; and a multi-variant missile (referred to as the “R2000” set) that included a hypersonic glide vehicle variant.
The actors used Claude Code in place of human software engineers to develop the guidance, navigation, and control (GNC) software that steers and stabilizes a flying vehicle. For example, they used Claude to integrate an open-source autopilot onto a phone-class flight computer, writing the control and position estimation software, tuning the control settings, running a firmware build pipeline, and performing a flight simulation. The actors managed several Claude instances at once, assigning each one a role, much as a lead would delegate work on a small engineering team: the actors tasked one instance with writing the code, another with research, and a third with reviewing the code the first instance produced.
Our safeguards blocked many of their requests, but not all of them. The actors used a variety of tactics to evade our safeguards, including hiding their goals and the products the software was meant for, and they split their work across multiple sessions so no single session revealed their full intent.
These actors carried out a sustained effort to develop guided weapons, including using Claude to design guidance software. We do not have evidence the actors succeeded in fielding an operational device; but they did test-fire a guided rocket. This field test appears to have failed: within hours, the actors returned to Claude to work out why it failed.
Expect more of this. AI systems democratize expertise and capability. Most of the time that’s a good thing, but sometimes it’s not.
Login
