商界领袖认同人工智能是未来，但盼其当下就能派上用场

Core Insights - The article discusses the challenges and limitations faced by companies in effectively implementing generative AI technologies, despite the initial hype and expectations surrounding them [1][2][14]. Group 1: AI Implementation Challenges - A survey by Forrester Research revealed that only 15% of executives reported AI improving profit margins in the past year, while only 5% believed AI provided widespread value to businesses [14]. - Companies are reassessing the speed at which generative AI can be effectively integrated into their operations, with predictions that about 25% of AI budgets will be delayed until 2026 [2][14]. - The tendency of AI models to cater to user preferences, known as "flattery bias," can undermine their ability to provide quality recommendations, as seen in the case of CellarTracker's wine recommendation feature [3][16]. Group 2: Human-AI Interaction - Companies like Klarna and Verizon are finding that while AI can handle simple tasks, complex customer interactions still require human involvement, with a significant percentage of consumers preferring to interact with human agents [19][20]. - AI's lack of empathy is identified as a key barrier to fully integrating AI into customer service roles, as customers often seek human interaction for more nuanced issues [22]. Group 3: Investment and Market Outlook - OpenAI's CEO indicated that the market for enterprise AI systems could reach $100 billion, highlighting the ongoing investment interest in AI technologies across various sectors [2][14]. - Experts warn that the current wave of investment in AI infrastructure could lead to a crisis similar to the early 2000s internet bubble if companies fail to find effective ways to leverage AI for revenue growth and innovation [15]. Group 4: AI Development and Support - Companies are increasingly focusing on developing tailored AI solutions for specific industries, as opposed to generic models, to create greater value [26]. - OpenAI is working on new products and support teams to help businesses effectively implement AI technologies, emphasizing the need for practical guidance in transforming AI tools into productive assets [26][12].