Some ways AI could kill us all

I don't think this is how it will actually play out. If you play a chess grandmaster, you can predict that they will beat you even if you can't predict how. I chose these examples because I don't think they require much imagination or accepting exotic assumptions. It is important to note that if chimpanzees were to guess how humans would decimate them, they would get it wrong. Chimpanzees would not imagine guns. They would not foresee poison gas. They would not conceive of chemical castration. They would not imagine humans going around and intentionally infecting them with AIDS. They have no concept of these things; they would not see it coming. Perhaps they might guess we'd be really good at throwing rocks. Amazingly good. Well, technically, that's what guns do: throw "rocks" really really well. So how will superintelligent AI actually wipe us all out? Probably in a way I couldn't conceive of. Nonetheless, it's not hard to see how deadly they could be with what we already know about. Method 1: engineer the deadliest and most contagious virus ever seen Coronavirus-19, aka COVID, looms large in the memory of living adults today. It started in December 2019 and within three months it had spread to 114 countries[1] and according to one estimate, infected 3.4 billion people by November 2021[2]. Before vaccines, COVID killed 0.5-1% of people infected, that is one out of every 100-200 people. Perhaps you think that's not that bad. It's hardly extinction, and no, it's not, but COVID is neither the most contagious nor the most deadly disease humanity has seen so far. Ebola kills 25-90% of those infected[3] ; smallpox killed 30%[4] of the unvaccinated, and the figure was 10% for SARS[5]. Rabies is 100% fatal. Measles, one of the most contagious diseases ever, is 4-6x more contagious than COVID[6]. COVID itself mutated over time and became twice as contagious as it was originally[7]. The Black Death killed tens of millions in Europe, and by historical estimates, that was 30-60% of Europe's population. Granted, they did have worse hygiene back then. Yet all of these diseases were naturally occurring. Blind, unthinking evolution produced them: tiny little packages of DNA that get into an infected host's cells and use the host's own cells to reproduce and spread further. Had we not encountered viruses for millennia, they would sound like science fiction. We have no reason to think that naturally occurring viruses are the worst that could be created. In fact, human researchers have already experimented with making existing viruses worse[8]. In 2012, researchers took H5N1 bird flu and modified it so that it could spread through the air, which it could not do before[9]. Importantly, contagiousness and lethality are not mutually exclusive. Biological trade-offs can constrain transmission, but there is no general rule that a more contagious pathogen must be less deadly. The extremes in these charts don’t necessarily combine freely—but neither does high contagiousness guarantee low lethality.[10] What could a superintelligent AI create, if it were trying on purpose? AI has already made major contributions to biology. In 2024, AlphaFold 2’s developers shared the Nobel Prize in Chemistry for predicting proteins’ three-dimensional structures from their amino-acid sequences[11]—a problem scientists had pursued for over fifty years. A protein’s shape helps determine what it does, and the ability to predict that paves the way for effective alterations. AI systems have also designed entirely new proteins that scientists manufactured and confirmed could perform intended functions. These achievements do not mean today’s AI can necessarily design a pandemic virus[12], but they are reason to believe that future superintelligent AI would be able to do this. Edit: on August 6, 2026, the Financial Times reported that AI has already been used to create synthetic viruses[13]. The tool used to create them, Evo 2, is open source and free for anyone to download. That's method number one: engineered super-pandemic. Method 1a: Engineer a pandemic against the food supply In the early 1990s, Uganda experienced an epidemic of cassava mosaic virus that affected 80% of the country's cassava-growing areas. Annual harvests fell from 3.5 million tonnes to as little as 0.5 million leading to 3000 deaths due to famine.[14] There were similar localized famines in Uganda.[15] Again, those were naturally occurring viruses. But once you are able to engineer viruses, you are not limited to targeting humans directly. A superintelligent AI could create additional viruses that attack plants and animals creating widespread famine. Misconception A: AIs don't have bodies, they can't act in the real world We humans live in the real world. AIs live in computers. At first, we might wonder how they could get out of the computers and be able to hurt us. Robotics is here Tens of thousands of highly capable robots have already been manufactured. Superintelligent AI co