Is it Time to Panic about American Ignorance of Deep Learning?
Eric Schmidt in a recent Verge report has a few remarks about the current state of AI policy in the United States:
He said that America needs to “get [its] act together as a country” to develop an AI strategy that involves both government and private industry. That way, he said, the US could expect to continue its leadership in developing AI. “Weren’t we the ones that invented this stuff? Weren’t we the ones that were going to go exploit the benefits of all this technology for betterment and American exceptionalism, in our own arrogant view,” he said, adding: “Trust me, these Chinese people are good.”
I’ve remarked previously (see: “Sputnik event of AI”) on what seems to be a lack of urgency in America with regards to the exponential progress in Deep Learning. Asian counties, ever since 2015, have the pedal to the metal in funding AI. Jeffrey Ding a student of Oxford’s Future of Humanity Institute has written a recent analysis of the Chinese government’s plans in AI. The Chinese government has an ambitious goal of dominating AI by the 2020s.
If one thinks that this is just the normal market and technology competition, then you are gravely mistaken. If American policy doesn’t correct itself soon, we will be what Cuban calls “SOL”:
Those building up a robotics industry here, investing in our AI industry here, that’s the new infrastructure. Because if we don’t do it, and China or Russia win those wars, we’re SOL. We’re out of luck, right?
China is already deploying its ‘social credit’ system to control its citizenry. This is a system that monitors a citizen’s social behavior using multiple data streams and assigns a ‘social credit’ that influences what you may or may not be able to access (i.e., travel, financial products, career etc.):
To gain a bit of an understanding of the state of AI in China, here is Kai-Fu Lee providing his own overview:
Where he says that the U.S.-China duopoly in A.I. has already arrived! It’s not some future scenario; it is here now.
There are plenty of indicators that China is deploying significant resources in this space. A recent report covers a $2.1 billion investment to build an “AI District”:
You can find more billion-dollar initiatives coming out of China. When we ask what’s happening in the USA, you can barely find $100,000 SBIR funding involving Deep Learning.
Just to emphasize the point even further, China’s president Xi Jinping appears to show an advanced command of the subject. This is revealed by the analysis of his bookshelf in a recent address to his nation. The Shanghaiist.com writes about this in “What’s new on Xi Jinping’s bookshelf this year”:
He is reading texts on understanding AI, AR, algorithms, and machine learning, including The Master Algorithm by Pedro Domingos and Augmented by Brett King.
What’s displayed on his bookshelf, of course, could just be all for show. However, is it not better to display that you know something rather than the alternative that you are completely clueless and proud of it? China is not subtle about its intentions in the AI arena. Pretty soon they will be flexing their muscles displaying their utter dominance. They are telegraphing their moves, and it appears that there is nothing that America is doing in response.
I usually use Google Trends as a tool to discover what people are interested in this world. What is alarming is the relative interest in countries searching for the term “Deep Learning”:
The United States is 30 times less interested in the term “Deep Learning” as compared to China. The top five countries are all in Asia (if you consider Israel to also be part of Asia). Three of the countries in the top five are countries (or cities) with populations less than 8 million people. That is 40 times smaller than the United States but with more than twice the interest. Despite the exponential and mind blogging progress in the area of Deep Learning, there is simply little interest among the American population.
Could this be perhaps because Americans have been fed a steady diet of Terminator and the Matrix scenarios by Hollywood? Is AI just one of those ‘inconvenient truths’ that we want to wish away as something to not worry about in the imminent future? Could it be the pervasive mind-set of the American citizenry thinking that the government is an unnecessary waste contributing to this apathy?
Scandinavians’ in stark contrast have the perception of A.I. as being something that will contribute to their greater well being. Is this due to the firm belief in these countries that the government is aligned with their own needs and beliefs?
The Japanese also perceive A.I. entirely different from Americans. The Japanese have always been enamored with robots. It’s not just prevalent in the factories, but also their entertainment. Japanese equivalent of super heroes is essentially robots or androids. It all began with Astro Boy in the 1950s and ever since then, the Japanese have been treated with a steady diet of Robots in their entertainment. The Transformers and Pacific Rim are two examples of this Robot meme that has invaded American theaters. The super heroes of the Japanese are not humans that gained mutant powers; they are humans that are in symbiosis with robotic skeletons.
Contrast the difference. “Ghost in a Shell”, a Japanese story where a human is embedded within a robot skeleton. This is a human that has been made to believe that she is a robot. The human becomes super-human by becoming a robot. Compare that with “Blade Runner 2049”, a biological robot that is trying to determine if it has the same intrinsic characteristics as a true human. The American psyche doesn’t appreciate the human that fuses with AI capabilities. AIs are distinct sentient forms that are always positioned as a threat to humanity.
It indeed is interesting that the Japanese relationship with AI and Robotics is at an emotional level. In a Time article “How Japan’s Radically Different Approach to AI Could Lead to Wild New Tech” writes:
In his book The Buddha in the Robot, Mori offers a perspective on human-automobile interaction that seems radical by western standards. In Mori’s view, humans and robots have a connection that is more egalitarian than master-slave.
There is a significant disconnect among the American population regarding the benefits of Artificial Intelligence (and Deep Learning as its primary driver). This disconnect appears to me rooted in American culture and American individualism. There’s a disconnect in many American’s notions of the purpose of government. Government is looked at as an unnecessary bureaucracy and an instrument of taxation. From this view point, the idea of creating AI as a common good (similarly as how the environment is a common good) is a foreign idea. Furthermore, the relationship of Americans with technology is about a master-slave relationship. Therefore any sufficiently intelligent automation is viewed as being something that flips that relationship around. Today we may be the masters, but we could become the slaves.
In summary, there are deep-rooted cultural issues that Americans have in relationship with Artificial Intelligence. These cultural biases may have a profound effect on future AI competitiveness. The Chinese, Scandinavians and the Japanese don’t seem to be carrying this same psychological baggage. These countries look at A.I. as a means to become a better self or to create a better society. Meanwhile, in the U.S.A. it is a zero-sum mindset that any investment in AI takes more jobs away. This is the same mindset that investments in renewal energy will take away the job of a coal miner.
One has eventually has to realize that change is all but inevitable. Intelligence, according to the late Stephen Hawking, is a measure of our ability to adapt to change. The AI driven economy is here today and it’s best that we prepare ourselves to adapt to this new world by becoming less ignorant. It is now time to truly begin to panic and become urgently to become more informed:
Update: The French are beginning to panic: https://www.aiforhumanity.fr. Good for them! https://www.reuters.com/article/us-france-tech/france-to-spend-1-8-billion-on-ai-to-compete-with-u-s-china-idUSKBN1H51XP
