Over the past ten years, futurist, NYU professor, and award-winning author Amy Webb has done a lot of study on people and groups related to AI. She says, “We have reached a fever in all AI cases.” Let us step back and see where this goes.
Her new book, The Big Nine: How the Tech Titans and their Thinking Machines Could Warp Humanity, is about this. It takes a quick look at the trends that, in her opinion, have put the progress of technology on a dangerous road. The “G-MAFIA” (Google, Microsoft, Amazon, Facebook, IBM, and Apple) in the US is limited in the short term by the constant needs of a capitalist market. This makes it hard to plan for AI in the long term with care. In China, Tencent, Alibaba, and Baidu are working together to get huge amounts of data that will help the government reach its authoritarian goals.
Webb says that if we do not change this path, we could instantly reach a disaster. But we can still do something, and everyone has a part to play. Her story was told to MIT Technology Review, along with her ideas on what we can do to make her feel better.
The Q&A below has been summed up and changed to make it clearer.
You say that worrying technical, political, and economic trends are coming together right now. Could you say more about what the tech trends are?
If you talk to a researcher in the field, they will tell you that many of the things that people say AI will do will not happen for a very long time. For example, vehicles that drive themselves, systems that can recognize everything, and AGI systems that can think more like humans will be a long time off.
From my point of view, the idea that we have some kind of walking, talking machine or a machine with a voice that makes choices on its own by looking at the horizon for the day is a bit off. There are already billions of small changes happening all the time that get worse over time and lead to systems that can make a lot of decisions on their own at the same time.
For instance, the DeepMind team has been putting in a lot of work to show computers how to beat people at games. A lot of success has been made in areas like learning with multiple tasks and hierarchical learning from reinforcements. AlphaZero, the company’s newest form of the AlphaGo algorithm, can learn how to play three games at once without any help from a person. That is a big change. There is also a brand-new area of study called “generative aggressive networks.” In this area, you can now make human faces from a lot of pictures that look incredibly real.
We have been told that these new technologies are more interesting and sexy than AGI. But if you look at things from 40,000 feet up, you can see that systems will decide what to do for us. We need to stop and think about what would happen if those systems let something we do not know happen instead of human plan.
How about the changes in the economy and government? Could you explain the ones that bother you the most?
People can share their thoughts freely in the United States. In this way, Silicon Valley got its start. Competition and new ideas have grown because of it. That is how AI and other technologies came to be.
But here in the US, too, we have a terrible lack of forethought. The federal government has stopped funding scientific and technical study instead of coming up with a big plan for AI or our long-term futures. So, the business sector is where the money must come from. They also expect a certain return, though. That is not good. You can not plan your R&D breakthroughs when you are working on study and important technologies. It would be great if the big tech companies did not have to keep planning conferences to show off their newest and coolest products. They work very hard as it is. Instead, we now have a lot of cases of bad choices made by someone in the G-MAFIA, most likely because they had to work quickly. We can start to see the bad results of the conflict between doing study that helps people and making investors happy.
That is already bad enough, right? But all of this is going on at the same time that China is combining a huge amount of energy. The sovereign wealth fund of China is mostly used for basic AI study. A lot of money is spent on AI. And when it comes to privacy and data, they do not think at all like we do. This means they have a lot more information that they can use to make things better. Since there is one central authority, it is very simple for the government to use data from 1.3 billion people to try and build AI services. That is just in their own country.
The Belt and Road Initiative looks like a normal building program, but it is also partly digital. It is not enough to just build roads and bridges; you also need to set up 5G networks, fiber optics, and mine and clean data in other countries. Some people who care about things like free speech and Western democratic values should not use these technologies.
As a country, why should we try to be more like the free West?
That is a great question. I have lived in the US, China, and Japan. And you might ask yourself, “Is that really the worst part?” when you see how bad things are in our country and China. For Americans, China’s social scoring system sounds crazy and awful. But what many people do not know is that self-reporting and watching how people act in villages and communities has always been a part of Chinese society. The social credit score just does that for you. That is a good question.
Imagine a perfect version of Western democracy next to a perfect version of Chinese communism. Which would you choose? I think Western democracy is better because it gives people more chances to succeed and for ideas to run freely. As an individual, I believe that rewarding good behavior is a great way to improve society and help each of us reach our full potential.
It is not fair to compare them now, with the way AI is taking the world. Do we have to look at the best and worst of both Western democracy and Chinese communism whenever we compare them?
What a great question! You could say that some parts of the AI environment are already having a bad effect on our democratic values in the West. The events that took place on Facebook clearly show us how things should be done. But also check out what is going on with people who are against vaccines. They spread completely false information about science and vaccines. Freedom of speech, platforms are platforms, and we have to let people say what they want. These are American values. Something that is wrong with this is that computers pick editorial content that makes people make bad choices and kids get sick.
The problem is that while our technology keeps getting better, our ideas about what free speech is and what a free market economy looks like have not changed that much. Most of the time, we take very simple interpretations: “free speech” just means that all speech is free, as long as it does not break defamation rules. The story does not end there. We need to have a more mature and smart talk about our current laws, our new technologies, and how we can make them work together.
To put it another way, you think that we will change from where we are now to a better example of Western democracy. And you would rather have that than the perfect communism in China.
Of course, I am sure it is possible. Everything I worry about is that everyone waits, that we drag our feet, and that it will take a real disaster to get people to move, as if where we are now was not already a disaster. However, the fact that measles is back in Washington state is terrible. That is also what took place after the election. Today, I can not see how the current political environment could be good for our future, no matter what side of the aisle you are on.
Because of this, I am sure that there is a way forward. But we need to work together to get Silicon Valley and Washington, DC on the same page.
What do you think the government, businesses, colleges, and regular people should do?
The process of AI growth is a problem, and we are all interested in it. You, me, my dad, my friend, and the man I walk by at Starbucks right now. So what should regular people do? Watch out for who and how they use your information. Look into what smart people have written for a short time to understand what we are talking about. Take some time to learn about sharing pictures of your kids before you start drawing attention to your life. It is okay to know what it means and what it might mean in the future. But you should know that first.
People in business and investors can not expect them to rush a product over and over again. It makes things hard for us on the road. Their attempts to be more inclusive and make sure their staff is more like people in the real world will need to be greatly increased so that they can do things like learn how to hire people. They can also stop. Any money or time put into an AI company, project, or whatever else it is must also be used to keep an eye on things like risk and bias.
Universities need to make room in their plans for hybrid degrees. They should push CS students to take classes in comparative literature, world faiths, microeconomics, cultural anthropology, and other related fields. They need to fight for programs that let people get two degrees at the same time in computer science and international relations, religion, political science, philosophy, public health, education, and other fields. People should not just learn ethics as a one-time thing to cross off a list. Even better, schools should push teachers to include deep conversations about bias, risk, philosophy, religion, gender, and ethics in their classes.
One of the most important things I think should happen is the creation of GAIA, which I call the Global Alliance on Intelligence Augmentation. Different groups of people around the world have very different ideas about how to collect and share data, what tasks can and should be automated, and what the future of smarter systems might be like. So, I believe we should create a main group that can create global rules and guidelines, a type of crash barriers that do not just let American or Chinese ideas into AI systems, but also allow much broader views of the world.
Above all, we need to be ready to think about this for a lot longer than five years. “Well, we can not know what will happen, so let us not worry about it now” has to stop being an excuse. We really can not tell what will happen in the future. We can plan better, though.
