Part I: Rise of the AI Tutor

第一部分：AI 导师的崛起

A great teacher can teach calculus with a paper clip and literature in an empty field. Technology is just another tool, not a destination.

—Unknown

一位伟大的教师可以用回形针教微积分，在空旷的田野上教文学。技术只是另一个工具，而不是目的。

—无名氏

In your Primer you have a resource that will make you highly educated, but it will never make you intelligent. That comes from life. Your life up to this point has given you all the experience you need to be intelligent, but you have to think about those experiences. If you don’t think about them, you’ll be psychologically unwell. If you do think about them, you will become not merely educated but intelligent.

—Neal Stephenson, The Diamond Age

在你的启蒙书中，你拥有一个可以让你受过高等教育的资源，但它永远不会使你变得聪明。聪明来自生活。到目前为止，你的生活给了你所有需要的经验来变得聪明，但你必须思考这些经验。如果你不思考它们，你将在心理上不健康。如果你思考它们，你不仅会变得受过教育，还会变得聪明。

—尼尔·斯蒂芬森，《钻石时代》

Throwing Away the Bottle

抛弃瓶子

The genie was out of the bottle. As quickly as ChatGPT seemed to spread to all reaches of the globe, the application faced widespread bans and resistance.

精灵已经从瓶子里出来了。ChatGPT 似乎迅速传播到全球各地，同时也面临着广泛的禁令和抵制。

OpenAI, the company that developed ChatGPT, had introduced a broad tool that allows for conversations and research assistance on a wide range of topics, but also for what many people consider cheating on student assignments and exams. In early 2023, the Los Angeles Unified School District became the first major school system to ban it. Seattle Public Schools were next, prohibiting generative AI on all campus devices. The district said it did not allow cheating and required original thought and work from students. From there, New York City Public Schools, the largest school district in the country, temporarily banned ChatGPT over the fear that students were using it to write essays and fill in answers to homework for them. Besides, it said, the tool did not help build critical thinking and problem-solving skills. Then Fairfax County Public Schools in Virginia banned it, followed by Alabama’s Montgomery County.

开发 ChatGPT 的公司 OpenAI 推出了一种广泛的工具，可以在各种话题上进行对话和研究协助，但也可以用于许多人认为在学生作业和考试中作弊的行为。2023 年初，洛杉矶联合学区成为第一个禁止使用 ChatGPT 的大型校区。接下来是西雅图公立学校，禁止在所有校园设备上使用生成式 AI。该学区表示不允许作弊，要求学生的原创思维和作品。从那时起，美国最大的学校系统纽约市公立学校暂时禁止了 ChatGPT，担心学生使用它来写文章和填写作业答案。此外，它还表示，这种工具不有助于培养批判性思维和解决问题的能力。然后是弗吉尼亚州的费尔法克斯县公立学校也禁止了它，接着是阿拉巴马州的蒙哥马利县也紧随其后。

OpenAI’s artificial intelligence chatbot had been released to the public the previous November, and in less than a week it had surpassed one million users. People were using it for things like answering questions, creating code, and writing essays, touting it as the next leap in technological innovation. Two months later, schools everywhere, from France to India to Australia, had outlawed the bot. Some equated the spread of the technology to COVID and pronounced it the death of education for our kids as we knew it. “Today we are facing a new sort of plague, one that threatens our minds more than our bodies. ChatGPT, the artificial intelligence chat bot that can write college-level essays, is going viral,” said one Inside Higher Ed opinion piece. “To their shock and dismay, [teachers] will find that their classroom has tested positive for GPT.”

OpenAI 的人工智能聊天机器人于前一年的 11 月公开发布，在不到一周的时间内用户超过了一百万。人们使用它来回答问题、创建代码和写文章，并将其吹捧为技术创新的下一个飞跃。两个月后，从法国到印度到澳大利亚的各地学校都禁止了这款机器人。一些人将这一技术的传播比作新冠病毒，并宣布它是我们所知的孩子教育的终结。一篇《高等教育内部》的评论文章称：“今天我们面临一种新型瘟疫，它威胁着我们的心智胜过我们的身体。ChatGPT，这种可以写大学水平论文的人工智能聊天机器人正在病毒式传播。”“令他们震惊和沮丧的是，[教师们]会发现他们的课堂已检测出 GPT 阳性。”

Frankly, as both a father and an educator, I understood this distrust. The last thing I wanted was for a new technology to come and strip our students of agency, creativity, socialization skills, and collaborative learning opportunities. Student exposure to generative AI is inevitable at this point, so it is natural to worry about the implications it will have on learning and development. Some say kids get enough screen time as it is, and we fear ChatGPT and other AI-based applications will only mean more screen time. We fear large language models will worsen instances of students not doing their own work. We fear that the potential effect on student writing skills will be catastrophic, with generative AI able to generate text quickly and efficiently for them. We fear that because GPT-generated text pulls from millions of online sources full of biased language and viewpoints, the information it lays out in its final form will be biased as well.

坦率地说，作为一名父亲和教育者，我理解这种不信任。我最不希望看到的是新技术剥夺学生的自主性、创造力、社交技能和协作学习机会。学生接触生成式 AI 在这个时候是不可避免的，所以担心它对学习和发展的影响是很自然的。一些人说孩子们已经有足够的屏幕时间了，我们担心 ChatGPT 和其他基于 AI 的应用程序只会意味着更多的屏幕时间。我们担心大语言模型会加剧学生不自己做作业的情况。我们担心对学生写作技能的潜在影响会是灾难性的，因为生成式 AI 可以快速高效地为他们生成文本。我们担心，因为 GPT 生成的文本来自数百万个充满偏见语言和观点的在线来源，其最终形式的信息也会是有偏见的。

These are all reasonable fears. However, I have always said that when it comes to technology and education, it is not that technology is good or bad, it is how you use it that matters. Yes, technology can suck you into unhealthy habits. It can make you pay more attention to notifications on your phone than to the people in the room with you. You can easily lose a few hours to social media, only to feel more insecure or triggered as a result. It is not hard to find very dark and disturbing content if you accidentally type the wrong word into search (and sometimes even the right word). That very same technology, though, also allows you to stay in touch with friends and family. Whether editing video or writing or coding, it can be a boon for your creativity and a powerful means of self-expression. And, closest to my heart, it can be a means to learn and improve yourself. Ideally, we use technology in the education space to enhance social connections, emotional development, and character.

这些都是合理的担忧。然而，我一直说，当涉及到技术和教育时，重要的不是技术本身是好是坏，而是你如何使用它。是的，技术可以让你陷入不健康的习惯。它可以让你比起关注房间里的人的注意力更多地关注手机上的通知。你可以很容易地在社交媒体上浪费几个小时，结果只会感到更不安全或被触发。如果你不小心输入错误的单词进行搜索（有时即使是正确的单词），很容易找到非常黑暗和令人不安的内容。然而，同样的技术也允许你与朋友和家人保持联系。无论是编辑视频、写作还是编程，它都可以成为你创造力的福音和强有力的自我表达手段。而且，最让我感到欣慰的是，它可以成为学习和提升自己的手段。理想情况下，我们在教育领域使用技术来增强社会联系、情感发展和品格。

We know that the most recent generation of AI can be incredibly powerful. In medicine, AI can assist in diagnosing diseases, analyzing medical records, and providing personalized treatment recommendations. Businesses are using large language models to streamline content generation and automate workflows. In legal and compliance spheres, large language models assist in contract analysis and legal research, generating documents, and ensuring adherence to regulations. From developing technical documentation, to writing user manuals, to creating grant proposals to coding, employing this technology is far more productive than avoiding it. The meme circulating on the internet that reads “You won’t be replaced by an AI, but you might be replaced by someone using AI” has some real truth to it.

我们知道，最新一代的 AI 可以非常强大。在医学领域，AI 可以协助诊断疾病、分析病历和提供个性化治疗建议。企业正在使用大语言模型来简化内容生成和自动化工作流程。在法律和合规领域，大语言模型协助合同分析和法律研究，生成文件，并确保遵守法规。从开发技术文档、编写用户手册、创建资助提案到编程，使用这项技术比回避它要高效得多。互联网上流传的一个梗“你不会被 AI 取代，但你可能会被使用 AI 的人取代”有一定的道理。

The most successful students will be those who use AI to help make conceptual connections for developing ideas. Students who learn to use AI ethically and productively may learn not only at an exponentially higher rate than others but also in a way that allows them to remain competitive throughout their careers. They will have a deeper understanding of the given subject matter, because they will know how to get their questions answered. Rather than atrophying, their curiosity muscle will be strengthened.

最成功的学生将是那些利用 AI 帮助建立概念联系以发展想法的学生。学会以伦理和高效方式使用 AI 的学生不仅可以以指数级的速度学习，还可以使他们在整个职业生涯中保持竞争力。他们会对给定的学科有更深入的理解，因为他们知道如何找到问题的答案。他们的好奇心不仅不会萎缩，反而会得到加强。

All these skills will directly translate into the workplace too. Those who can steer AI to partner with them, and know what great writing entails, will be those who get the best output from the technology. Those who feed their curiosity will be the most likely to see around corners in the marketplace. Those who can brainstorm with the AI and their colleagues will appear more creative than those who don’t use AI or those who completely outsource their work to it. Workers will need to learn how to use large language models to automate almost any traditional white-collar process, too, from collating information to doing analysis on spreadsheets.

所有这些技能也会直接转化到职场。那些能够引导 AI 与他们合作并知道优秀写作需要什么的人，将是从技术中获得最佳输出的人。那些激发自己好奇心的人最有可能在市场上预见到转折点。那些能够与 AI 和同事进行集思广益的人将比不使用 AI 或完全将工作外包给 AI 的人显得更有创造力。员工还需要学习如何使用大语言模型来自动化几乎所有传统的白领流程，从整理信息到分析电子表格。

I am not going to say that there are no issues that come with this new tool. Greg Brockman, OpenAI’s president who first demoed GPT-4 to me in the summer of 2022, believes that the challenge of identifying and tackling these issues comes down to putting safety measures in place. “From the start, when OpenAI developed GPT-1, safety was the most important problem to solve. As we push this technology forward, we want it to be beneficial and we want it to be safe,” he says. The company has invested significant resources in implementing safeguards to prevent misuse of the tool, from mechanisms that prevent the AI from sharing knowledge about illegal activities, to blocking disallowed content, to personal data safety. When it comes to artificial intelligence and merging it with our kids’ education, these types of guardrails are even more important. The work involved can be labor intensive, but the cause is well worth it, he says. “Fundamentally, signing up to make the most important technology that humans will ever create, and applying it to education, is something to get behind.”

我不会说这种新工具没有任何问题。OpenAI 总裁格雷格·布洛克曼在 2022 年夏天首次向我演示 GPT-4，他认为识别和解决这些问题的挑战在于实施安全措施。“从一开始，当 OpenAI 开发 GPT-1 时，安全是最重要的问题。随着我们推进这项技术，我们希望它是有益的，我们希望它是安全的，”他说。公司投入了大量资源实施保障措施，以防止滥用该工具，从防止 AI 分享非法活动知识的机制，到屏蔽不允许的内容，再到个人数据安全。当涉及到人工智能并将其与孩子们的教育相结合时，这些类型的防护措施变得更加重要。参与的工作可能非常繁重，但这个原因是值得的，他说。“从根本上说，致力于制造人类有史以来最重要的技术，并将其应用于教育，是值得支持的事情。”

More significantly, he says, the technology is something not to fear but to use. Engagement with generative AI can potentially benefit students by providing new ways for them to learn the material, enhancing collaborative learning, stoking creativity, promoting socialization, and even helping kids through mental health issues, as well as offering new ways for parents and teachers to engage more thoroughly with their children’s education. Rather than pushing our kids to avoid AI, Brockman says, we can help them learn smarter.

更重要的是，他说，这种技术不是用来害怕的，而是用来使用的。与生成式 AI 的接触可以通过提供新的学习材料的方式来潜在地使学生受益，增强协作学习，激发创造力，促进社交，甚至帮助孩子们解决心理健康问题，还提供了父母和教师更深入参与孩子教育的新方法。布洛克曼说，与其让我们的孩子避开 AI，不如帮助他们更聪明地学习。

The genie is out of the bottle. It is time to throw the bottle away and our fear of generative AI along with it.

精灵已经从瓶子里出来了。现在是时候扔掉瓶子，以及我们对生成式 AI 的恐惧了。

“点赞有美意，赞赏是鼓励”