soros122_ARIS MESSINISAFP via Getty Images_ukrainewar

Politics

民主制度能在多重危机下幸存吗？

Jun 6, 2023 George Soros

发自纽约——我们生活在混乱的时代，太多事情发生得太快了，让人们感到困惑不已。为此哥伦比亚大学经济历史学家亚当·图兹（Adam Tooze）发明了一个广为传播的词来形容这种状况：“多重危机”。

多重危机存在许多成因。在我看来困扰当今世界的多重危机的首要成因是人工智能，其次是气候变化，俄罗斯对乌克兰的入侵位列第三。当然还有很多其他成因，但我会集中讨论这三点，希望能有助于减少困惑。

人工智能

当微软通过一家名为OpenAI的关联公司将ChatGPT免费提供给公众时，人工智能震惊了世界。那是在2022年11月。ChatGPT对谷歌的商业模式构成了存续性威胁，为此后者力争尽快发布一个产品来与之竞争。

To continue reading, register now.

Subscribe now for unlimited access to everything PS has to offer.

As a registered user, you can enjoy more PS content every month – for free.

Already have an account? Log in

Introductory Offer: Save 30% on PS Digital

Subscribe now to gain greater access to Project Syndicate – including every commentary and our entire On Point suite of subscriber-exclusive content – starting at less than $5 per month.

Subscribe Now

George Soros

Writing for PS since 1997
126 Commentaries

George Soros is Chairman of Soros Fund Management and the Open Society Foundations. A pioneer of the hedge-fund industry, he is the author of many books, including The Alchemy of Finance, The New Paradigm for Financial Markets: The Credit Crisis of 2008 and What it Means, and The Tragedy of the European Union: Disintegration or Revival? His most recent book is In Defense of Open Society (Public Affairs, 2019).

new comment has been posted. new comments have been posted.

0 Comments on this paragraph, 25 Comments on this article

Kacper from Warsaw Jun 19, 2023

I must admit that I do not understand the threat of AI (as a threat independent from its makers) because, as something void of any actual understanding, it's hard to imagine AI actually outsmarting humans. It can seem very smart, but, as mentioned, it still seems to be (and it doesn't seem that it will seize to be) very vulnerable to exceeding human stupidity (hallucinations). AI is fundmentally different than humans, and hence is good at what humans are not (complicated computations, fast writing) and bad at what humana are good (actual understanding, e.g. Captcha).
A new reply to this comment has been posted.
Richard Warth Jun 16, 2023

Mr Soros has long colluded and profiteered from the cash from chaos world of international finance, that relies on minimal regulation or accountability. To now be in a moral dilemma over Western society's inability to protect itself from the Trojan horse of AI is akin to another of Goethe's works, Faust. The moral being you cannot sell your soul to the devil, then expect events to somehow result in Entente Cordiale.
A new reply to this comment has been posted.
Luc Lefebvre Jun 14, 2023

"in an open society, the role of the state is to defend the freedom of the individual; in a closed society, the role of the individual is to serve the interests of the rulers". Joseph Henrich in "The Weirdest People in the World" shows that things are more complicated and deep-rooted than that. We better take his analyses into account to avoid a bipolar confrontation between two worlds based on too simplistic analyses probably on both sides.
A new reply to this comment has been posted.
Anton Kristoff Jun 12, 2023

Oh yes, climate change is such a joke...even the Antarctic ice sheets are cracking up.
No, I thank the writer for summing up the challenges humanity faces (and for his positive outlook in the last paragraph). Illiberal "democrats" lacking in understanding and global responsibility should not distract us from accepting the severity of these challenges.
Because of its urgent and all-embracing nature, I would put the climate issue first, without belittling the other two (Russia's particularly barbaric conduct of the war is causing great harm to both the climate and the ecology of the planet, i.e. the water from the Dnepr now flooding the wheat fields, Nordstream releasing gigantic amounts of climate gases while attacks on civilian structures will divert much needed human energy to reconstruction).
So we must do our best to prevent wars from taking place in the future (it's a terrible indictment of humanity's present stage of development that nuclear dissuasion still plays a part). I am grateful to a writer who as an impeccable analyst of financial crises (among others) decided to address and convey the urgency of the overriding challenges on these pages and at the Munich Security Conference this year. but unfortunately, it has turned out that the Coronavirus pandemic was easier to grasp and deal with for people than global warming.
I am more optimistic about the continued attraction of "open societies" and the demise of populism. Europe, for example, is sticking together on Ukraine. In any case, we do need to examine our way of life and change entrenched habits. It will keep us on our toes. If some commentators find climate warnings all too alarmist, let's point out to them: It's not just many scientists, but other decent people, young and old, who sound this alarm. People who are not known for particularly extreme views. Among them the secretary-general of the United Nations, Antonio Guterres.
A new reply to this comment has been posted.
Rene Leclezio Jun 12, 2023

Of the 3, Putin worries me the most. As Sting’s song goes: I hope the Russians love their children too.
A new reply to this comment has been posted.
Vidvuds Beldavs Jun 8, 2023

For now AI is a rapidly growing set of tools and capabilities that can be used to address the cascading crises now facing humankind.
With partners in Africa we are advancing the African Smart Communities project to empower communities with knowledge to address critical problems including lack of access to electricity and clean drinking water. We plan to build interfaces and translators to enable use of ChatGPT and other tools via conventional mobile phones even in communities with high illiteracy that are monolingual in vernacular languages.
Climate change is growing challenge for rural communities in sub-Saharan Africa where about 600 million people do not have access to electricity.
Russia's war against Ukraine is highly similar to the attempt by ISIS to conquer Iraq and Syria and then step by step gain control of the Muslim world. The attempt by Russia to recreate a mythic Russian world world of the past will fail because it is a delusion perpetrated by an authoritarian state rather than a democratic state with individuals and communities empowered empowered by knowledge to protect and build their society cooperatively with like-minded people around the world.
The extraordinary environmental damage being caused by Russia's war may be exceeded in cost by the damage that Russia's war is causing to science and the use of science and facts to address real world problems. Over 2000 schools in Ukraine have been destroyed. World class laboratories in the Donetsk, Kharkiv and other areas impacted by the war have been damaged and looted by Russian military.
ISIS did not breach the Mosul dam when it was in control of that part of Iraq. The Kakhovka dam has been breached as a result of Russia's war against Ukraine.

Global science has been heavily damaged by Russia's war against Ukraine. CERN and other major international scientific initiatives have seen the exit of Russian scientists. At a time when every thinking mind is needed to address climate change, pandemics, and other problems Russia exploits the pandemic to launch its war against Ukraine. The present Russian leadership is trying to create a Russian world that never existed with no heed to global crises. As a result of Russia's actions global capacity to address climate change, pandemics and other challenges is being dramatically diminished.

Russia needs to come to its senses to join with the rest of the world to address global crises rather than to use global crises to advance its goal of a creating a Russia World that ignores climate change, pandemics and other global challenges.

The new authoritarianism that is anti-human rights, anti-gay, with limitations on women's right resonates with people that cannot adapt to change. Knowledge, including knowledge empowered communities that effectively use AI tools can prevail against authoritarian states because of the power of knowledge that enables innovative responses to crises.
A new reply to this comment has been posted.
Barry Rosenfeld Jun 8, 2023

Circa 1966, an episode on Star Trek featured a character named Khan (played by Ricardo Montalban), the last survivor of the Age of Dictators that arose in the mid-21st Century. Have the writers of that episode been granted uncommon prescience? Democracy cannot deal with the climate crises (yes, a plural) because only authoritarian governments have the power to enforce the measures needed to forestall, or adapt to, a tipping point that might already have been reached.

The crux of the climate crises is the significant overpopulation of the planet. Indeed, economists foretell disaster if populations shrink at the same time as foretelling disaster if the consumption of carbon and carbon-based resources continues. AI, if used properly, might enable the substantial growth in productivity needed to offset the significant reduction in population that can only be done by an authoritarian government (Mau, where are you now that you are needed?)

In addition, only an authoritarian government can control the flood of immigration that is destabilizing the political environments in receiving countries. Only an authoritarian government, moreover, can control production, allocation and prices needed to provide for an environmentally sustainable political economy.

Finally, as Mr. Soros knows, the world is run already by a financial and corporate hoi aristoi, and it is only a matter of time that popular political support based on socio-cultural factors will provide the electoral basis for de jure authoritarian rule by those who control the world's wealth and resources.

Welcome to the Age of Dictators.

PS: Star Fleet was a military dictatorship. Does anyone remember any civilian political control?
- Neil MacLean Jun 8, 2023
  
  Well Turkey must be great to your way of thinking. How’s that working? Authoritarians aren’t well disposed to feedback which hinders effective adaptation. China had a period of effective and also before that pretty ineffective authoritarians. Xi to my mind seems to be trending poorly. Wolf warrior diplomacy may feel satisfying to execute but I doubt its long term effectiveness. China has demographic challenges that increasingly challenge its stability - not that the rest of us don’t have challenges too. Social media really screw up effective social feedback as truth has no currency - merely whatever crap motivates more screen time.
  - Barry Rosenfeld Jun 8, 2023
    
    Why the ad hominum response? Reread your last sentence. The issue for discussion is can authoritarian or democratically governed states deal with the challenges better? All you seem to say is that several authoritarian states have problems. No kidding! So do the democratically governed ones. BTW, what does Wolf Warrior "diplomacy" have to do with global warming? What would you call US diplomacy, e.g., foxy (Ukraine), ostrich-like (Israel/Palestine), hammer headed (Latin America), or harebrained (China)? As for global warming and climate change in general ... ?
    - Neil MacLean Jun 8, 2023
      
      Ad hominum response? Nothing personal said or intended. Basic point is impaired feedback in authoritarian government - it seems you missed that. Wolf warrior diplomacy is example of China being ineffective. Point about social media is how feedback in democracies is now impaired by the unconstrained greed of social media companies - you could say it is helping us evolve away from democracy the most critical characteristic of which is that government responds to feedback that is representative of society as a whole / that gets impaired when society becomes subject to mass delusion unconnected to their realities.
      A new reply to this comment has been posted.
    A new reply to this comment has been posted.
  A new reply to this comment has been posted.
A new reply to this comment has been posted.
Tibor Parniczky Jun 7, 2023

AI is processing information through algorithms designed by human beings.
AI went through a long development before it became a very effective tool and was made accessible to everyone for free. The concern is also real that there is no free lunch.
A new reply to this comment has been posted.
Neil MacLean Jun 7, 2023

I've made an effort to track down a number of different discussions with Prof Hinton and have listened carefully. I also have a far greater background in computing than Mr Soros or many other commentators. I tend to believe in the soundness of prof Hinton's understanding of how his algorithms incrementally improve themselves and their ultimate ability, granted sufficient computing power, data and time to finds ways to get around any obstacles we may present to thwarting the objectives being pursued by the AI. So I do buy what the author above is saying with respect to facing a crisis in uncontrolled AI evolution. And I also fully buy the crisis in climate change and think his 10 meter rise in sea level is a lot less than the 70 meters plus forecasts from worst case scenarios forecasters I tend to credit, being flabbergasted by the utter lack of progress being made, including by my country Canada. The ocean temperature can be a big factor in sea level rise from thermal expansion beyond just melting all naturally occurring ice. With respect to Ukraine I welcome some optimism but fear that attack is inherently more costly in lives than defense, the Russians are well dug in, and the Ukrainians have a far more limited population, with Putin having painted himself into a corner of no surrender.
A new reply to this comment has been posted.
Tibor Parniczky Jun 7, 2023

AI is a product of human invention. It is modeling human thinking, not even human nature or behavior - if one not asking to do so.
- Tibor Parniczky Jun 7, 2023
  
  So I agree that measures that work in other cases when human nature must be controlled should be applied. Transparency and accountability are good models, not just for societies and countries but for all AI tool producers and users.
  A new reply to this comment has been posted.
A new reply to this comment has been posted.
Robert Wolff Jun 7, 2023

AI is Not Dangerous – It is Human Gullibility that is Dangerous:

My attitude about AI has been that it is not dangerous in itself at all. The problem is uneducated people who have little understanding of reality and are therefore totally gullible to disinformation. George Soros explains, “AI is very good at producing disinformation and deep fakes and there will be many malicious actors.”

Clearly, there is no easy answer as to how to change human nature from being gullible to disinformation and following the directions of those who create it, even without AI. Climate Deniers and Covid Deniers and Election Deniers are just recent examples which prove that many, many people prefer to believe those who deny reality because it makes them “feel better”. It means they do not have to alter their ritual routines in the face of new threats to society just to avoid being the victims of these new threats. They simply pretend reality is irrelevant and continue to perform their dangerous daily rituals, ignoring real threats to themselves and others.

The bottom line is that we need to regulate AI because we cannot change the gullibility of human nature to prefer fictious misinformation over reality to guide their actions – when reality demands they change their ritual behaviors for themselves and humanity to survive.
- Omar Luquin Jun 8, 2023
  
  You nailed it!
  A new reply to this comment has been posted.
A new reply to this comment has been posted.
Tibor Parniczky Jun 7, 2023

Again, the question is how one defines smartness and intelligence. And yes, if you can define being smart and intelligent, you can devise an algorithm surpassing it.
The hint here is that AI will do it not at human request, but on its own. Now we should define "live" so that it could surpass its physical (hardware) limits.
- Tibor Parniczky Jun 7, 2023
  
  The point here is that this kind of bombastic argument would not help the cause. It is rather the example of a fourth crisis: the trusted communication crisis.
  A new reply to this comment has been posted.
A new reply to this comment has been posted.
Frank Youell Jun 7, 2023

Global life expectancy is at all-time highs. Global crop yields are rising. Global temperatures and ocean levels are slowly (quite slowly) rising. A global climate crisis? Not exactly.
A new reply to this comment has been posted.
Gennady Shkliarevsky Jun 7, 2023

I am not sure why Mr. Soros bothers to write. I guess it is because he owns this publication. Is he trying to play the role of public intellectual? It is too late for that. He obviously has nothing new to convey to the public. Just the old platitudes and tired theories. He just makes himself look out of date.
- Martin Edelson Jun 11, 2023
  
  I disagree. Mr. Soros has views to share and they are particularly valuable reflecting the views of someone who cares about people, the world, and the forces that shape our current reality. He is not out of touch and does not repeat arguments endlessly. His views need to be taken seriously by everyone, particularly those who can effect change in our society.
  A new reply to this comment has been posted.
A new reply to this comment has been posted.
Tibor Parniczky Jun 7, 2023

In support of sounding the bell, I would remain factual. The situation is similar to rationality in economics. From the beginning of economics making rational decisions has been a postulate. Simply saying that these systems are better might be easily questioned and not only by neurologists. What do we mean by "better"?
Processing more data concurrently faster and more coherently and consequently than we, who might be biased and tend to forget - indeed, it is possible. But both the samples which they process and the algorithms they use are human. Meaning, there might be one of us having the same information and arriving at the same conclusion as the machine.
A new reply to this comment has been posted.
Frank Youell Jun 6, 2023

Democracy is dying and the author has played a key role in killing it. In China, they don't put images of freaks on beer cans. In the US they do. A system that puts freaks on beer cans has no future. There is rather good quote from Camille Paglia that covers this.

She says that androgyny becomes prevalent “as a civilization is starting to unravel. You find it again and again and again in history.”
A new reply to this comment has been posted.
Peter McLoughlin Jun 6, 2023

Humanity has always been on a “narrow and winding path”. But unless it heeds the warning of history it will not avoid WW III, and grasps a simple syllogism: every empire in history has eventually faced the fate it was trying to avoid – its own demise; everyone wants to avoid nuclear annihilation; so, there will be a nuclear war. Paradoxically, unless that logic is accepted our fate is sealed.
https://patternofhistory.wordpress.com/
A new reply to this comment has been posted.

https://prosyn.org/Ztt78FHzh

民主制度能在多重危机下幸存吗？

人工智能

Introductory Offer: Save 30% on PS Digital

New Comment

Contact us

Log in/Register

民主制度能在多重危机下幸存吗？

人工智能

Introductory Offer: Save 30% on PS Digital

Edit Newsletter Preferences

Set up Notification