Core Viewpoint - The article discusses the potential for autonomous vehicles to achieve rapid deployment through advancements in technology, cost reduction, and the evolution of social rules [1][2]. Technology: From Ideal to Reality - Technological innovation is the fundamental driver of new developments, with a significant shift from content-based generative AI to goal-driven intelligent agent AI expected to lead to breakthroughs in autonomous driving capabilities [3]. - Two main technological approaches in autonomous driving are identified: "end-to-end" technology, which requires vast amounts of high-quality data for training, and modular technology, which combines human-designed algorithms with neural networks [4]. - Current autonomous driving systems primarily offer driver assistance rather than full autonomy, constrained by technological capabilities and costs [4]. Cost: From Niche to Popularity - Cost reduction is crucial for the commercialization and widespread adoption of new products, as seen historically with the introduction of the Ford Model T, which made cars affordable for the middle class [7]. - Significant advancements in AI cost reduction, particularly in China, are expected to drive explosive applications in autonomous driving, with examples like DeepSeek achieving training costs significantly lower than competitors [8]. - Companies like Tesla are actively working on reducing costs, with projections for autonomous taxi services to be economically viable by 2026 [8]. Rules: From Phenomenon to Institutional Framework - The integration of autonomous driving into society requires adaptive rules and regulations, as technology alone cannot address all challenges [10]. - Historical precedents show that technological advancements often lead to societal and cultural shifts, necessitating a reevaluation of existing norms and values [11]. - Establishing long-term rules for autonomous driving is essential, particularly concerning safety, responsibility allocation, and the ethical implications of AI decision-making [13][14].

Group 1: Technology - The advancement of AI technology is shifting from content generation to goal-driven intelligent agents, which is expected to lead to significant breakthroughs in autonomous driving capabilities [2] - Two main technological approaches in autonomous driving are identified: "end-to-end" technology, which requires vast amounts of high-quality data for training, and modular technology, which combines human-designed algorithms with neural networks [3][4] - Current autonomous driving systems are primarily in the realm of assisted driving rather than full autonomy, limited by technological capabilities and costs [4] Group 2: Cost - The reduction of costs is crucial for the widespread adoption of new technologies, as seen historically with the introduction of the Ford Model T, which made cars affordable for the middle class [5] - China has made significant progress in reducing AI training costs, exemplified by DeepSeek's training costs being one-thirtieth of OpenAI's, which may accelerate the application of autonomous driving [6] - Companies like Tesla are also focusing on cost reduction, with projections for autonomous taxi services to be economically viable by 2026 [6] Group 3: Regulation - The integration of autonomous driving into society requires adaptive regulations that reflect technological advancements and societal needs [7] - Historical precedents show that technological progress often leads to significant societal changes, necessitating a reevaluation of existing rules and norms [7] - Establishing foundational rules for autonomous driving, such as human-machine relationships and liability distribution, is essential for future industry development [8] Group 4: Safety - Research indicates that 90% of traffic accidents are caused by human error, and transitioning to algorithm-driven driving could reduce accidents significantly [9] - The ethical implications of autonomous driving decisions, particularly in unavoidable accident scenarios, highlight the need for societal consensus on moral choices [9] - Extensive testing is required to ensure the safety of autonomous vehicles, with estimates suggesting that they need to cover 440 million kilometers without errors to match human driver safety levels [10]