Summary of Key Points
This article discusses the transformation of innovation paradigms in the context of new quality productivity and the organizational changes driven by AI. The main arguments include:
1. The innovation paradigm has shifted from the "producer-centered" approach of the industrial era (the Schumpeter model) to a "user-centered" one in the digital age, with users becoming the source of innovation.
2. Physical AI (AI with a physical presence, such as robots and autonomous vehicles) is driving organizations to evolve from being machine-assisted to achieving human-machine collaboration and co-governance.
3. Companies need to address the governance challenges posed by algorithmic power, restructure the way they collaborate with talent, and utilize physical AI to enhance their global competitiveness in manufacturing.
1. Innovation Has Changed: From Experts Inventing to Users Taking the Lead
In the past, innovation was associated with universities, research institutes, or corporate R&D departments—such as engineers designing new smartphones, which follows the Schumpeter paradigm. However, things have changed in the digital age. Demands are diverse and rapid, and knowledge is distributed among ordinary people; many innovations actually come from users. For example, if you find a flaw in an app and suggest a improvement, and the company adopts it, that becomes a new feature. Similarly, home enthusiasts modifying computer hardware also contribute to innovation. Why has this happened? Because users understand their own needs better, and their small improvements in practical scenarios are often more practical than what experts come up with in laboratories. This not only aligns with the laws of the digital age but also reflects a "people-centered" approach—innovation is no longer the exclusive domain of a few elites; it involves hundreds of millions of ordinary people.
2. Physical AI Is Here: Organizations Moving from Machine Assistance to Human-Machine Cooperation
Physical AI refers to AI that can interact with the physical world, such as the Kiva robots in Amazon warehouses, Tesla's autonomous vehicles, and collaborative robots working alongside workers in BMW factories. Previously, AI was used for assistance (e.g., helping with report writing by finding information). Now, it plays a more cooperative role—robots assist in tasks like assembling parts, while humans make decisions and make adjustments. This change requires organizations to revise their rules: management's focus should no longer be on controlling people but on ensuring effective collaboration between humans and AI. Managers should act as facilitators rather than dictators, and employees and AI should be equal partners.
3. Is Algorithmic Power Too Great? Companies Need to Practice "Responsible Governance"
Platform companies are becoming increasingly powerful due to their algorithms—such as those used for delivery scheduling or product recommendations in e-commerce. These algorithms can be like "digital monsters" that determine who gets orders and which ads users see, potentially leading to unfair practices that traditional regulations cannot address. What needs to be done? Platforms must implement responsible governance by integrating external regulatory requirements into their internal policies, making algorithms understandable, controllable, and accountable. For example, they could establish specialized departments to regularly check for algorithmic issues to prevent them from getting out of control.
4. How Should Employees Collaborate with AI? Three Models Determine Who Will Win
In the AI era, whether employees can collaborate effectively with AI will directly affect a company's competitiveness. The article identifies three collaboration models:
1. Joint Design: Employees and AI work together to design workflows, where AI analyzes data and employees adjust strategies based on the results, creating value together.
2. Frequent but Extensive Use: Employees frequently use AI for repetitive tasks, such as generating content without making modifications, resulting in average outcomes.
3. Passive Adoption: Employees completely rely on AI, without using their own judgment (e.g., simply copying reports written by AI), which can lead to errors.
The article emphasizes that AI is not a leveler of abilities; employees who do not know how to use it will be marginalized, while those who do will become more competitive. Companies need to foster "AI leadership" by enabling their employees to both utilize AI effectively and make informed decisions.
5. The Upgrading of Chinese Manufacturing: From "World Factory" to "Industrial Foundation Model"
The impact of physical AI on manufacturing is not just about making machines smarter; it involves reshaping the fundamental logic of factories. The path for upgrading Chinese manufacturing includes three steps:
1. World Factory: China used to produce goods for others, such as assembling smartphones.
2. Factory of Factories: Now, China can provide production solutions to other countries, teaching them how to use AI robots.
3. Industrial Foundation Model: Similar to how ChatGPT is a foundational model for AI, China can develop industrial-specific "foundation models"—universal intelligent production systems that other companies can customize for their own production lines. These steps will transform Chinese manufacturing from being a mere producer of goods to a provider of advanced capabilities, reshaping the global trade landscape.
Conclusion
This article highlights that in the era of new quality productivity, innovation relies on users, organizations must adapt to human-machine collaboration, companies need to manage algorithms effectively, and employees should learn to work with AI. These changes are not just about technology but about how organizations and people interact. Those who can adapt first will be the ones to succeed in the future.