人工智能如何帮助降低大学成本

By 通过

Nov. 19, 2024 at 11:00 am ET
2024 年 11 月 19 日上午 11:00 东部时间

Artificial intelligence has the potential hold down prices in complex and expensive services such as higher education. Photo: Getty Images/iStockphoto
人工智能有潜力降低高等教育等复杂且昂贵服务的价格。照片：Getty Images/iStockphoto

Kartik Hosanagar (@khosanagar) is a professor of technology and digital business at the Wharton School of the University of Pennsylvania and faculty co-lead of AI for Business. He also is the author of “A Human’s Guide to Machine Intelligence: How Algorithms Are Shaping Our Lives and How We Can Stay in Control.”
卡尔蒂克·霍萨纳加（@khosanagar）是宾夕法尼亚大学沃顿商学院的技术与数字商业教授，同时也是商业人工智能项目的联合负责人。他还是《人类对机器智能的指南：算法如何塑造我们的生活以及我们如何保持控制》的作者。

Much of the discourse around artificial intelligence involves its potential to displace millions of jobs. But AI offers a surprising upside, too: It could make life cheaper, especially for many of the professionals who most fear it.
关于人工智能的讨论中，许多内容涉及其可能取代数百万个工作岗位。但人工智能也提供了一个令人惊讶的好处：它可能使生活成本降低，尤其是对于那些最害怕它的许多专业人士。

To be sure, I’m not talking about the prices of everyday necessities like eggs, bread or gas. In fact, AI might actually exacerbate price increases in those areas because of the increased energy demands associated with AI chips. Rather, it is more complex and more expensive services such as higher education, software and professional services where AI has the potential to hold down prices.
当然，我不是在谈论鸡蛋、面包或汽油等日常必需品的价格。事实上，由于与人工智能芯片相关的能源需求增加，人工智能可能会加剧这些领域的价格上涨。相反，人工智能在更复杂和更昂贵的服务领域，如高等教育、软件和专业服务，具有抑制价格的潜力。

The reason stems from AI’s ability to automate and augment tasks at a low cost. Historically, automation has driven down prices. During the 19th century, for instance, the introduction of automated machinery such as power looms and spinning machines made mass-produced fabrics more affordable for the general public. AI will have a similar impact. AI doesn’t just replace workers, it also can act as their co-pilot, taking on some of their tasks to clear their plate for others or adding to their capabilities. Whether AI replaces workers or boosts their productivity, the result will be higher margins for producers, lower prices for consumers, or both.
原因在于人工智能能够以低成本自动化和增强任务。历史上，自动化一直在降低价格。例如，在 19 世纪，自动化机械如动力织机和纺纱机的引入使得大规模生产的面料对公众更加实惠。人工智能将产生类似的影响。人工智能不仅仅是替代工人，它还可以作为他们的副驾驶，承担他们的一些任务，以便为其他任务腾出时间，或增强他们的能力。无论人工智能是替代工人还是提高他们的生产力，结果将是生产者的利润更高，消费者的价格更低，或者两者兼而有之。

WSJ 华尔街日报

My experience as a professor illustrates this relationship clearly. From 2004 to 2023, private university education costs in the U.S. rose by 40% when adjusted for inflation, with public university costs increasing even more steeply. Administrative costs were the primary driver of these increases, as universities spent more on customized student services such as career counseling, extracurricular opportunities and wellness checks, all of which require additional personnel.
我的教授经历清楚地说明了这种关系。从 2004 年到 2023 年，美国私立大学的教育成本在调整通货膨胀后上涨了 40%，而公立大学的成本上涨得更为陡峭。行政成本是这些增长的主要驱动因素，因为大学在定制学生服务上花费更多，例如职业咨询、课外机会和健康检查，这些都需要额外的人手。

AI can address these rising costs by enhancing the efficiency of student services. An AI system, for example, could analyze a student’s academic records and suggest personalized career paths, allowing career counselors to focus on providing tailored advice rather than data collection or routine counseling. Colleges might need fewer counselors as a result, helping bring down costs and helping control tuition increases. Similarly, AI could improve the productivity of educators by supporting tasks such as lesson-planning, setting homework and exams, and grading, allowing them to spend more time teaching.
人工智能可以通过提高学生服务的效率来应对这些不断上升的成本。例如，一个人工智能系统可以分析学生的学业记录并建议个性化的职业路径，使职业顾问能够专注于提供量身定制的建议，而不是数据收集或常规咨询。因此，大学可能需要更少的顾问，从而帮助降低成本并控制学费上涨。同样，人工智能可以通过支持课程规划、布置作业和考试以及评分等任务来提高教育工作者的生产力，使他们能够花更多时间进行教学。

There also is good reason to believe AI will drive down costs in other industries, such as enterprise software, where AI is enabling new players to challenge established companies by drastically reducing development costs. Similarly, there could be many applications for AI in healthcare—from drug discovery and precision medicine to AI-powered scribes that will allow physicians to spend more time with patients than in front of computers.
还有充分的理由相信，人工智能将降低其他行业的成本，例如企业软件，人工智能使新兴企业能够通过大幅降低开发成本来挑战成熟公司。同样，人工智能在医疗保健领域也可能有许多应用——从药物发现和精准医疗到人工智能驱动的记录员，这将使医生能够花更多时间与患者交流，而不是在电脑前。

It’s important to note that traditional economic indicators, such as gross domestic product, may fail to fully capture the value of a technology like AI. If AI transforms a previously paid-for service such as medical transcription into a free one, GDP’s focus on monetary transactions will miss the economic value created. As such, we likely will need new ways to fully measure the economic and societal value of AI. One approach is to measure changes in customer welfare as a measure of AI’s impact instead of focusing exclusively on monetary spending.
重要的是要注意，传统经济指标，如国内生产总值，可能无法完全捕捉像人工智能这样的技术的价值。如果人工智能将以前需要付费的服务（如医疗转录）转变为免费的服务，国内生产总值对货币交易的关注将会忽视所创造的经济价值。因此，我们可能需要新的方法来全面衡量人工智能的经济和社会价值。一种方法是衡量客户福利的变化，以此作为人工智能影响的衡量标准，而不是仅仅关注货币支出。

Write to Kartik Hosanagar at reports@wsj.com.
写信给 Kartik Hosanagar，邮箱地址是 reports@wsj.com。