The author of "A Brief History of Mankind" talks about AI: The 2028 US presidential campaign may no longer be hosted by humans

• 1 "Language is the operating system of human culture. From language came myths and laws, gods and money, art and science, friendship and nations, and computer code. AI's mastery of language means it can now hack into and manipulate civilization's operating system. ”
• 2 "Social media is AI's first contact with humans, and humans lose. Large language models are our second contact with AI, and we can't afford to lose any more. ”
• 3 "Imagine when you board an airplane and half of the engineers who built it tell you that there's a 10 percent chance the plane will crash, killing you and the rest of the plane. Will you still board the plane?
• 4 "In 2022, in a survey, more than 700 top academics and researchers from leading AI companies were asked about the risks of future AI. Half of respondents said there is a 10 percent or greater chance that future AI systems will cause humans to extinction (or permanent severe incapacitation). The tech companies that build today's big language models are caught up in a race to get all of humanity on that plane. ”

Recently, Yuval Harari, author of A Brief History of Mankind and Israeli historian, and Tristan Harris and Aza Raskin, producers of the well-known documentary "Surveillance Capitalism: The Social Dilemma," co-authored an article calling on humanity to buy time to upgrade existing institutions (which still date back to the 19th century) in response to the AI world. And learn to control humans before AI takes control of it.

Israeli historian Yuval Harari, author of A Brief History of Mankind.

The article, titled "You Can Have a Blue Pill or a Red Pill, We Don't Have a Blue Pill Anymore," was published in The New York Times. The red pill and blue pill is a popular meme that means choosing between the "red pill" of disturbing knowledge and harsh reality and the "blue pill" of remaining ignorant and continuing to live an ordinary life.

Without strict safety checks on products, drug companies cannot sell new drugs to people. Biotech labs can't release new viruses into the public domain in order to impress shareholders. Similarly, AI systems with GPT-4 and above capabilities should not be involved in the lives of billions of people faster than culture can safely absorb them. The race to dominate the market should not set the pace at which humanity's most important technology can be deployed.

"The specter of artificial intelligence has haunted humanity since the mid-20th century, but until recently it remained a distant prospect of something that belonged to science fiction rather than serious scientific and political debate. It is difficult for our human minds to grasp the new capabilities of GPT-4 and similar tools, and even harder to master the exponential speed at which these tools develop more advanced and powerful capabilities. But most key skills boil down to one thing: the ability to manipulate and generate language, whether with words, sounds, or images. The article reads.

"At first it was words. Language is the operating system of human culture. From language came myths and laws, gods and money, art and science, friendship and nations, and computer code. AI's mastery of language means it can now hack into and manipulate civilization's operating system. Through the mastery of language, artificial intelligence is seizing the master key of civilization, from the vault to the Holy Sepulchre. The article continues.

What does it mean for humans to live in such a world? "A world where a large percentage of stories, melodies, images, laws, policies, and tools are shaped by non-human intelligence that knows how to exploit the weaknesses, biases, and addictions of the human mind with superhuman efficiency, while knowing how to form intimate relationships with humans? In a game like chess, no one can hope to beat a computer. What happens when the same thing happens in the realm of art, politics, or religion? The author asked.

"AI can quickly eat the entire human culture — everything we've produced over thousands of years — digest it and start pouring out a flood of new cultural art. Not just school essays, but also political speeches, ideological manifestos, holy books for new cults. By 2028, U.S. presidential campaigns may no longer be hosted by humans. ”

The article argues that human beings often do not have direct contact with reality, but are shrouded in culture and experience reality through cultural prisms. "Our political views are formed by journalists' reports and anecdotes from friends. Our sexual preferences are regulated by art and religion. So far, this cultural cocoon has been woven by other humans. What would it be like to experience reality through a prism generated by non-human intelligence? ”

"For thousands of years, we humans have been living in the dreams of other humans. We worship gods, pursue beautiful ideals, and dedicate our lives to causes that stem from the imagination of some prophet, poet, or politician. Soon, we will also find ourselves living in the illusion of non-human intelligence. ”

For example, the "Terminator" series of movies depicts robots running and shooting in the streets. The Matrix assumes that to gain complete control over human society, AI will have to first gain physical control of our brains and connect them directly to computer networks. However, once it masters language, AI has everything it needs to control humans in a matrix-like hallucinatory world without having to shoot anyone or implant any chips in their brains. If a shot is necessary, AI can get a human to pull the trigger, just telling the right story to the human.

"Social media is AI's first contact with humans, and humans lose," the article states. The first contact gave us a taste of the bitter fruit that followed. In social media, the early AI was used not to create content, but to organize user-generated content. The AI behind news waterfalls is still choosing which words, sounds, and images reach our retinas and eardrums, based on selecting those that will get the widest spread, the most response, and the most engagement.

"While very primitive, the AI behind social media is enough to create illusions, exacerbate the polarization of society, destroy our mental health, and disintegrate democracy." Millions of people confuse these hallucinations with reality. The United States has the best information technology in history, but American citizens can no longer agree on who wins elections. Although everyone is now aware of the drawbacks of social media, the problem has not been solved because too many of our social, economic, and political institutions have become entangled with it.

The documentary Surveillance Capitalism: The Intelligence Trap reveals the consequences of humanity's growing reliance on social media.

"Large language models are our second contact with AI, and we can't afford to lose anymore. But on what basis should we believe in humanity's ability to align these new forms of AI with our interests? If we continue as usual, new AI capabilities will once again be used for profit and power, even if it inadvertently undermines the foundations of our society. ”

The article concludes by highlighting that AI does have the potential to help beat cancer, discover life-saving drugs, provide solutions to the climate and energy crises, and countless other unimaginable benefits. "However, if the foundation collapses, it doesn't matter how tall the skyscraper of AI-assembled benefits is."

The authors make an appeal: "It is time to take AI seriously before our politics, economics and daily lives become dependent on it." Democracy is a dialogue that depends on language, and when language itself is invaded, dialogue is interrupted and democracy becomes unsustainable. If we wait for chaos to happen, it will be too late to remedy it.

"We have summoned an alien intelligence. We don't know much about it, except that it is very powerful, providing us with dazzling gifts, but also undermining the foundations of our civilization. We call on world leaders to respond to this moment with a level commensurate with their challenges. The first step is to buy time to upgrade our 19th-century institutions for the AI world and learn to control it before AI takes control of us. ”

Yuval Harari has long been interested in the impact of artificial intelligence on the world, authoring books such as A Brief History of Humanity, A Brief History of the Future and A Brief History of Today, and founded Sapienship International Consulting. Tristan Harris and Aza Ruskin, founders of the Center for Humane Technology, produced a documentary called "Surveillance Capitalism: The Smart Trap," also translated as "The Social Trap," revealing the consequences of humanity's growing reliance on social media: society is facing a life-and-death crisis, dominated by artificial intelligence, big data and algorithmic models, leading to the fragmentation and chaos of recent years.