Amuums who insist on Amnesty International will kill us all

Translation Bible destruction It is published by the prophets of extinction of artificial intelligence, Yardkovsky and Nate Suares later this month, is “why all the supernatural intelligence will kill us all.” But “why all the supernatural intelligence will kill us,” because even the participating authors do not believe that the world will take the necessary measures to prevent artificial intelligence from eliminating all non -superheroes. The book goes beyond darkness, as it reads like notes full of observation in a litigant prisoner on the night before the execution of dawn. When I meet the self -appointed Cassandras, I ask them explicitly if they believed that they would personally meet their purposes through some cancellation tools. The answers come immediately: “Yes” and “Yes”.

I am not surprised, because I read the book – the title, by the way, is If anyone built it, everyone dies. However, it is the abyss to hear this. It is something, for example, he writes about cancer statistics and completely another to talk about reconciliation with a deadly diagnosis. I ask them how they think the end will come to them. Yaudkovsky initially evades the answer. “I don’t spend much time filming Zawali, because it does not seem a useful mental idea to deal with the problem,” he says. Under pressure, he bets. “I think suddenly fall,” he says. “If you want an easy version, something about the size of mosquitoes or maybe a dust mites fell on the back of my neck, that is.”

The technical aspects of a fictional, fantastic blow to the dust mites in which artificial intelligence works, and Yudowsky believes that it is worth the problem of knowing how to do this. Perhaps he could not understand it anyway. Part of the central argument of the book is that SuperismyLigence will come with scientific things that we cannot understand more than people can imagine from accurate treatments. Soares the co -author also says he imagines that the same will happen to him, but he adds that, like Yudkowsky, he does not spend much time to live on the details of his disappearance.

We do not stand a chance

The reluctance of the perception of the circumstances of their personal demise is a strange thing to hear from the people who composed a full book on everyone demise. For lovers-porn fans, If anyone built it It is reading the appointment. After joining the book, I understand the ambiguity of the domination of the way in which artificial intelligence ends our lives and all human life after that. The authors speculate a little. Boiling oceans? Sunnah prevented? All guesses are likely to be wrong, because we are being held in the mentality of 2025, and artificial intelligence will think of the future.

Yudkowsky is the most famous apostate in artificial intelligence, as it turns from the researcher to Reaper years ago. He even finished TED talk. After years of public debate, he and his co -author enjoy an answer to every counter -argument that was launched against their terrible encouragement. For beginners, it may seem intuitive to be numbered by LLMS, which often stumbles on the simple account. Do not be deceived, as the authors say. “AIS will not remain stupid forever,” they write. If you think excellent artificial intelligence will respect the boundaries that humans draw, they forget them, as they say. Once the models start teaching themselves to get more intelligent, AIS will develop “preferences” on its own that will not comply with what we want to prefer. In the end they will not need us. They will not be interested in us as chat partners or even as pets. We will be a source of inconvenience, and they will cancel the elimination of us.

The battle will not be fair. They believe that artificial intelligence initially may need human help to build its factories and its factories – which are easily done by stealing money and bribery people to help it. After that, it will build things that we cannot understand, and these things will end. “One way or another,” write these authors, “The world fades to black.”

The authors believe that the book is a kind of trauma treatment to remove humanity from its consent and adopt the sharp measures necessary to stop this unimaginable conclusion. “I expect to die of this,” says Suares. “But the battle did not even end up already.” It is very unfortunate, then, that the solutions they suggest to stop the destruction appear to be far more than the idea that the software will kill us all. Everything is summarized in this: press the brakes. Monitor data centers to ensure that they do not sponsor the cancellation process. She bombing those who do not follow the rules. He stopped publishing the papers with ideas to accelerate a march to experts. Did they have banned, ask them, 2017 paper on transformers Which started the movement of artificial intelligence. Oh yes, they will be, they respond. Instead of chat-GPT, they want Ciao-GPT. Good luck stood this industry trillion dollars.

Play

Personally, I do not see his own nucleus before a bite in the neck with some advanced dust. Even after reading this book, I don’t think it is likely to kill us all artificial intelligence. Yudksowky has already been involved Harry Potter FantasyAnd fictional extinction scenarios in which it revolves is very strange so that my human mind cannot be accepted. I think that even if Superintligence wants to get rid of us, he will stumble at the age of genocide. Amnesty International may be able to hit humans in a fight, but I will bet on it in a battle with the Murphy Law.

However, the disaster theory does not look impossibleEspecially since no one really sets a roof how to become smart intelligence. Studies also indicate Think about blackmail To avoid re -training, in one experience. It is also concerned that some researchers who spend their lives in building and improving artificial intelligence believe that there is an unsuccessful opportunity that can happen worse. One survey indicated Nearly half of the artificial intelligence scientists who respond, have linked the possibilities that species calm 10 percent or above. If they think that, it is crazy to go to work every day to achieve Aji.

My blaming told me that Yudkowsky and Soares Spin scenarios are very strange so that they could not be real. But I cannot be certainly They are wrong. Every author dreams that their book is classic. Not so much. If they are right, there will be no one about reading their book in the future. Just a lot of decomposing bodies that one day felt a slight spot at the back of their necks, and the rest was silent.