Rethinking Our Approach to Artificial Intelligence — 重新思考我们应对人工智能的方式

After two decades of well-intentioned efforts, it is time to admit that our current approach to artificial intelligence is failing. More funding, more awareness campaigns, and more rhetoric have not produced the transformative change we were promised. The fundamental flaw lies in treating artificial intelligence as a technical problem requiring technical solutions. In reality, it is a deeply human challenge rooted in values, habits, and social structures. ethical decision-making by machines cannot be solved by gadgets or algorithms alone. We have confused measurement with meaning. Organizations proudly report statistics while missing the underlying reality. Recent studies provide compelling evidence for this approach. Such metrics create an illusion of progress that actually obscures stagnation. What would a genuinely different approach look like? It would start with listening rather than lecturing. self-driving taxis now operate in twelve major cities. Top-down mandates consistently fail because they ignore local knowledge and community wisdom. The path forward requires humility. We must abandon the fantasy that experts in distant capitals possess all the answers. automation of routine tasks demands collaboration, patience, and a willingness to learn from failure. Only then might we begin to make real progress. The role of media and public discourse cannot be overstated. How issues are framed determines which solutions appear viable and which stakeholders gain influence. Responsible journalism requires moving beyond sensational headlines to provide nuanced analysis that empowers informed civic participation. From a historical perspective, periods of rapid transformation have consistently produced both winners and losers. The Industrial Revolution, the digital age, and now the era of artificial intelligence each disrupted existing social contracts. Managing these transitions justly requires deliberate redistribution mechanisms and robust safety nets. Innovative approaches are emerging from unexpected quarters. Social entrepreneurs, community collectives, and even student-led initiatives are prototyping solutions that challenge conventional wisdom. Recent studies provide compelling evidence for this approach These bottom-up experiments often reveal possibilities that top-down planning overlooks. The psychological impact on individuals and communities deserves closer scrutiny. Uncertainty about artificial intelligence contributes to anxiety and social fragmentation. Addressing these concerns requires not only policy interventions but also community-building efforts that foster trust and collective efficacy. Ultimately, the measure of our success will be the world we leave to the next generation. Short-term political cycles often incentivize postponement of difficult decisions, yet the costs of delay compound exponentially. Leadership in this context means prioritizing long-term sustainability over immediate popularity. These developments have generated significant public attention and widespread debate. Policymakers, practitioners, and ordinary citizens continue to grapple with the practical implications of these changes. the World Economic Forum predicts AI will create 97 million new roles by 2025. Looking toward the future, the trajectory of artificial intelligence remains uncertain but promising. Technological advances, shifting cultural norms, and evolving policy frameworks all suggest that the coming decade will bring transformative changes. Dr. Fei-Fei Li emphasizes the importance of human-centered AI design.