News

Recently, I had the privilege to attend the inaugural AI Expo for National Competitiveness in Washington DC representing Planetary Systems AI. The AI Expo is the place to convene and build relationships around AI, technology, and U.S. and allied competitiveness. The AI Expo is meant to serve as a forum for industry, government, and academic research entities to exhibit some of the latest technological breakthroughs — in AI, biotech, energy, networks, compute, microelectronics, manufacturing, augmented reality, and beyond — and discuss their implications for U.S. and allied competitiveness.

One of the many sessions I attended over the span of two days included Chairman Eric Schmidt’s fireside chat with Foreign Affairs’ Dan Kurtz-Phelan. Here are some key takeaways from that conversation about innovation, AI, technology, allied partnerships and more.

The capacity for Innovation, the speed of domination has long been central in global power. At the national level, however, would be a public competition. Suppose you argue that this time, is it? It’s not just the ingredients of economic or military or cultural power that there’s something distinctly on point in non-military non-asset security areas.

Soft power vs hard power:

The most valuable companies in the world are all now tech companies. That should be an indication that what is happening is much bigger than a single company or your project or your whatever you’re doing. Schmidt’s argument in the current debate on innovation, which came out of conversations at AI Expo, was that the traditional framing and our policy and political science, is car power as an example in the ban of Chinese EVs (electric vehicles) in the United States. Hard power is weapons and soft power is essentially culture and influence. However, the way you’re going to win is by the speed with which you can innovate, whether it’s in the hard power business of weapons or the soft power machine through innovation and competition in AI. The reason this is happening is that tech industry is building platforms that are very different and global in nature when it comes to United States and China. Other examples of soft power include Hollywood, YouTube, and TikTok, which Schmidt refers to as being similar to television media and resources controlled by China. Whether you debate this is good or bad, this is innovation power.

“Innovation and scale around innovation at scale”:

Another example of innovation power is found through the courage and resistance of what is happening in hard conflict zones. In Ukraine in contrast with a much larger adversary and area, innovation is used to hold on and for a while to buy time now, illustrating varied amazing hard power (weapons) combined with innovation power against defense. So we’re going to see whether they are going to make and at least and really hold its scale. Speed of innovation is critical in advanced technologies.

China and AI Development:

China is becoming a dominant player in the electrical field, through solar power and manufacturing. It is not something Schmidt would have predicted, but it’s obviously a part of their industrial policy which differs from America’s.

China will lead in surveillance and will be very good at that, he said: They will ultimately have to regulate their AI systems heavily to keep to their rules about speech and freedom. The Chinese government will likely shut down the speech and activity that would constrain innovation on AI development front. This is an advantage that democracies have, minimal constraints on innovation.

AI is the platform for everything and, according to Schmidt, China is well behind for a number of reasons including:

1. English is the most common training model language: The current generation of schools of models are trained using programming language. There is a lot more English language than the various Chinese languages on the web, including data to train models, but China is working on closing the gap.
2. Enter chain thought reasoning:  when a model has been trained enough where you can say what you want and the computer can write Python code or another command to make it happen. When AI can mimic human behavior or thought reasoning, then it begins to appear like general intelligence.
3. Access to compute chips: Trump and Biden Administrations limited access to chips essentially above NVIDIA A100 relationships and with other chip manufacturers, so the latest upgrades in compute chips are not available in China, which makes it much harder for the Chinese tech companies to reconfigure and stay competitive.

4. Different AI industrial models. China is trying to make money with AI apps and monetizing on application revenue models, which can result in a higher probability of success given how well apps have performed globally. For platforms out there where AI applications are being subscribed to in America, the business models have been done with huge venture investment of millions of dollars. That is an amount of money that is not available in the Chinese system. So the Chinese ecosystem is a “field hamstrung by all of these obstacles” and the government is trying to figure how to fix that.

 

AI supply chain China 

There was a decoupling in semiconductor products, but trade with China went up. U.S. and China are like Siamese twins – one body, two heads, depending on each other, but don’t always get along. Co-dependence is for global safety, but the CHIPS Act is a good thing.

EU and global AI innovation dynamic:

Schmidt says that Europe has an EU AI Act, which makes it almost impossible to innovate at large scale, guaranteeing that the approach Europe is taking limits people and they are not going to compete in the strategic power game when it comes to AI. The development of non-human intelligence that makes our world better is the game for the rest of our lives along with the China question. He suspects that China will have trouble with this. The UK government started, and the U.S government followed. He predicts that Korea and France will follow in the next four months.

AI Talent & Pipeline

Africa

When asked about Africa and the rate of entrepreneurship and innovation happening on the continent, Schmidt points out one thing that is restraining innovation – the lack of access to technical universities is inhibiting progress. When he was CEO of Alphabet, he had a rule of placing at least 10 people into a country in Africa. People love Google in Africa. Why? There is a lack of available textbooks, so Google was used to teach science students, where technical learning occurs, and help graduate students. There are many natural resources in Africa for AI manufacturing, but Schmidt does not think Africa will likely invent AI.

America (doesn’t) want you:

With regards to researchers and technologists, America’s strength is immigration. Schmidt and many business leaders agree that Washington’s stupidest policy is take the smartest people in the world, educate top minds at the best schools and then kick them out and create a competitor to America. It is not a good strategy – stopping STEM talent or math and science people who want to work on competitiveness. (i.e., TSMC)

China’s Quantum Program is open to people/immigrants who have been kicked out of America and that talent drain and innovation potential and top STEM skills are transferred.

Other suggestions from Eric Schmidt in multi-lateral diplomacy discussions include:

No Surprises Rule when it comes to AI, communicate ahead of time any risks or issues. This will take a long time to negotiate.

Shared Society Risks

U.S., China, EU, etc have shared risks in nuclear weapons, which is to be avoided at all costs.

The late Henry Kissinger, who was a strong proponent of starting conversations with China now; The strategic power competition between U.S. and China is present, so you have to when diplomacy and soft power can be leveraged and avoid hard power and war as much as possible. What war does to young men and women is so brutal. War is to be avoided at all costs.

Other speakers included Palantir’s Alex Karp and CTO Shyam Sankar as well as representatives from big tech (NVIDIA, Cerebras, Microsoft, Meta, etc.), U.S. government agencies and government representatives along with small businesses in the commercial/private industry sectors.