Schmidt first candidly pointed out Google's dilemma in the AI field.
He bluntly stated that Google focused too much on employees' work-life balance rather than fully committing to AI competition, which led to their struggle in competing with companies like OpenAI and Anthropic. He emphasized, how can a company maintain a leading position in AI if employees only come to the office one day a week?
"Look at Musk, look at TSMC, these companies are successful because they can push their employees. You must pressure your employees enough to win. TSMC will have physics PhDs work in factories in their first year. Can you imagine American PhD students on assembly lines?"
Schmidt's view directly addresses the profound reflection on the balance between work intensity and efficiency in the current tech industry.
In recent years, tech and internet companies have been reflecting on whether free lunches and dinners, high-welfare taxi fares, and luxurious gyms have really increased work efficiency.
Next, Schmidt highly praised Microsoft's collaboration with OpenAI. He believed that Microsoft's decision to outsource AI business to OpenAI was visionary. This collaboration model not only accelerated AI technology innovation but also helped Microsoft gain an advantageous position in the AI field.
In contrast, Apple's performance in the AI field seemed too conservative and slow, reflecting the bureaucratization and decision-making delays of large enterprises when facing emerging technologies.
Schmidt also mentioned the inspiration that TikTok's rise brought to the United States. He pointed out that in the entrepreneurial process, one should dare to take risks and innovate. If you succeed, you'll have money to hire top lawyers to clean up after you; if you don't succeed, no one will sue you.
When talking about OpenAI, Schmidt revealed OpenAI's huge funding needs for the Stargate project and predicted that this project might require far more than $100 billion in funding support.
Schmidt pointed out that the United States' leading position in AI is not unshakeable and needs continuous investment to maintain competitiveness.
Developing the AI industry requires enormous investments in electricity and funding.
Regarding the strategic debate on whether AI should be open-source or closed-source, Schmidt believes that the open-source model is indeed effective, but because AI investment is a financial bottomless pit, companies with open-source models find it difficult to maintain operations in the long term.
Schmidt also expressed disappointment with Europe's performance in the AI field.
He believes that Europe has always lacked sufficient investment and determination in technological innovation, causing them to fall far behind the United States in AI competition. In comparison, while France shows some potential, Germany and other European countries seem to be struggling.
"Brussels (the location of EU headquarters) has always been destroying opportunities for technological innovation."
Finally, Schmidt made predictions about the future development of AI.
He believes that AI will bring about a profound revolution, not only changing our way of life and work but also reshaping the global economic and political landscape.
"AI will make the rich richer and the poor poorer, and the same goes for countries. This is a game between strong nations. Countries without technological resources need to get a ticket to join the supply chain of strong nations, otherwise they will miss out on the feast."
Schmidt believes that AI now is like electricity was in the past - valuable, but still requiring organizational innovation to truly reap huge returns. Currently, everyone is still just picking the "low-hanging fruit."
Schmidt's speech provides us with a comprehensive perspective to examine AI development.
Overall, the AI industry is at a critical stage of development, full of opportunities but also facing many challenges.
Capital and tech giants are frantically "throwing money," "throwing models," "throwing computing power," "throwing data," and even "throwing registrations." But apart from bringing historically high points of AI frenzy and tech stock prices, the world hasn't changed much, and super applications that can trigger "mass participation" are still nowhere to be seen.
What can users do with AI? How can enterprises use AI to reduce costs and increase efficiency? These core issues are still being explored.
We still need to continue to address these challenges through constant innovation and organizational change to promote the healthy development of the AI industry.