科技與未來／AI、教育與人性困境─大學生自省

在過去一年裡，我時常思考用人工智慧（AI）輔助課業所涉及的倫理問題。和許多同儕一樣，我在大學期間開始使用ChatGPT，有時用來激發靈感或修改論文，有時則在課業壓力山大時用來節省時間。

我不認為在教育中使用人工智慧本質上是違反道德的，但隨著我愈來愈依賴它，我開始更深入地反思科技與教育和人性的關係。特別是對照自己學習中文的經驗後，我逐漸得出一個更深刻的認知：即雖然使用AI並非不道德的行為，但我們不能毫無顧忌地依賴它。就像許多技術產物一樣，這需要審慎的判斷：既非全盤否定，亦非盲目接受。

改變想法的瞬間

回顧過去，有兩個關鍵的瞬間塑造了我對智慧科技、效率與教育的看法。

第一個瞬間來自高中時代，一名才華洋溢的同窗以思想深度與文字功底著稱，作業永遠被老師當作範文，從未有過作弊行為。某日課堂上她隨口說出一句令我異常震驚的話：「也許作弊並不像教育體系宣稱的那般不道德，或許只是對現行體制的本能反應。」她並非為學術不端辯護，而是指出我們這些「優等生」和完美主義者始終承受著追求完美的巨大壓力。正是這種壓力，而非惰性，有時會讓人企圖走捷徑。

我認為她道出了真相。我就讀的高中學術上的競爭相當激烈，大學入學率幾乎是唯一指標。正是那些為92分與97分之差痛哭的完美主義者們更可能違背規矩。

對此我也不例外，我會使用翻譯軟體完成外語作業，與朋友核對答案。但智慧科技的特殊之處在於：與傳統作弊工具不同，它不僅沒有加劇我的完美主義傾向，反而治癒了這種病態心理。

緩解「完美」焦慮

最初我用AI潤飾語法、精選詞彙或調整論調，但最近我用它完成一些更高深的工作：總結文章的核心內容，找出閱讀材料的共同點，甚至起草文章的一部分。雖然我還未使用AI撰寫整篇論文，但我懷疑這其中唯一的原因是我只有免費版本，比不上付費版本的強大功能。我有時想AI已經成為一種心理依賴。但頗具諷刺意味的是，這種依賴反而緩解了我的完美焦慮。人工智慧幫助我在次要領域達到「及格線」，從而為真正熱愛的學科領域騰出時間和精力。

中文正是我真正感興趣的領域，從初中系統學習直至如今大學兩年過去，即便沒有學分要求，我仍持續選修相關課程。大學期間，當同時應付兼職、課業與心理問題時，我會用智慧工具加速中文寫作。但對掌握這門語言的熱忱從未消退。多年來，我努力內化的不僅是文法規則與字彙量，更是方塊字裡流淌的歷史血脈與文化基因。我在喬治城大學通過了兩門高級漢語的考試，滿足了我的專業對外語的要求，但我感覺自己還未熟練這門語言，部分原因可能是我依賴人工智慧完成一些作業。為了提高我的中文水平，我暑假期間在北京參加了一個高強度的中文學習班。

這個決定映射出AI最終教會我的一個道理：如何明確我真正的愛。它透過分擔機械性課業負擔，讓我更能全神貫注於我真正在乎的領域，例如外語學習。它幫助我分辨什麼是應付課業，什麼是有意義地學習。

協助「價值澄清」

這種「價值澄清」的過程讓我想起大一政治哲學課討論的內容。教授闡釋馬克思預言的「後稀缺社會」——當生產力突破某個臨界點，勞動將不再是生存必需，人們只為熱愛而工作。

這個觀點深植我心，巧合的是我當時正初次嘗試用智慧工具輔助寫論文。

坦白說，那次成果拙劣，我還不明白如何設計有效的提示詞，但從此愛上了人工智慧。我開始思考：人工智慧會否成為通往馬克思烏托邦的階梯？是否能幫助我們最終實現勞動解放，能否讓我們重新發現那些使人之所以為人的特質？

這個問題在我其他選修的課程中也不斷出現，從中文課的地域認同到哲學研討課上關於自我認知建設的討論。

在這些討論中，我逐漸意識到我們的自我認知總是相對於其他事物而言的。在此意義上，人工智慧幫助我更了解人性。

隨著我們愈來愈多利用機器完成那些重複性高的學術任務，我們愈能更清楚地認識到人工智慧無法達到的境界：它們無法復刻人類的好奇心、情感紋理、倫理辨識，以及個人的聲音。諷刺的是，透過與人工智慧的對比，我們開始更清楚地認識到學習過程的哪些部分更人性化，而且更不可被取代。

最近在中國短期留學時的一次偶遇再次印證了這個認知。我和一名年輕的教師聊天，雖然我們並非相熟，交流也不帶特殊情感，但卻給我留下了深刻印象。

依賴與恐懼並存

我們討論了教育、文化和經濟壓力，我們意識到中美兩國都面臨著一個困境：學生們都經常使用人工智慧協助完成作業，同時他們又擔憂人工智慧將在未來取代他們的工作。就業市場在緊縮，兩國的年輕人，不管是上海還是紐約，都感受到就業的壓力。

那名老師談到的學生對人工智慧一方面過度依賴，一方面又深感恐懼的狀況，與我的美國同學驚人地相似。我也提到，在美國對就業前景的焦慮存在很大的城鄉差別，生活在農村的大學生傾向於抱怨海外競爭，而城市裡的學生更多地認為，對他們來之不易的學位形成主要威脅的將是人工智慧。

這次對談並沒有讓我們找到何去何從的解決方案，但卻讓我更確信，擺在我們面前的挑戰並不是完全抵制人工智能，而是如何巧妙地使用它：明辨何時它能拓展認知邊界，何時會禁錮思維活力，何時又將誘導我們迷失本心。

短期內人類不可能也不該放棄使用智慧工具，但必須保持清醒的自我拷問：它在教育中和生活中的作用到底是什麼？對我而言，答案永遠指向求知欲、熱情，以及人類最深層的需求，對自己和別人的理解，即使我們身處的這個世界，以及我們探索這個世界所需的工具，都在無時無刻地發生變化。

AI, Education, and the Human Dilemma:A Student’s Perspective

In the past year, I’ve found myself returning often to a central question: What are the ethics of using AI to assist with schoolwork? Like many of my peers, I began using ChatGPT in college—initially to brainstorm ideas or revise essays, and occasionally to save time when I was overwhelmed with responsibilities.

This usage did not feel inherently unethical, but the more I relied on it, the more I began to interrogate my relationship with the technology. Especially when I contrasted my use of AI with my experience learning Chinese, I came to a more complex conclusion: using AI in education is not immoral by default—but it is not a tool to be used unthinkingly. Like many technologies, it requires discernment, not rejection or blind acceptance.

Two moments have shaped how I think about AI, productivity, and learning. The first occurred in high school during a conversation with a classmate widely respected for her integrity, intelligence, and writing skills. She once remarked, “Maybe cheating isn’t as immoral as the school system says it is. Maybe it’s just a reaction to the system we’re in.”She was not endorsing dishonesty. Instead, she was pointing to the crushing pressure students often feel—especially high-achieving, perfectionist students—to meet impossible standards. Her argument was that it’s not laziness, but fear of failure, that sometimes leads students to cut corners.

I attended an academically intense high school where college admissions were the ultimate goal. Ironically, it was often the most conscientious students—the ones who cried over a 92 instead of a 97—who were most likely to bend the rules. I was not immune to that pressure. I used Google Translate for language homework; I double-checked assignments with classmates. But what makes my current relationship to AI more nuanced is this: unlike those tools, AI has not amplified my perfectionism. It has, in some ways, helped me manage it.

Initially, I used AI to check grammar, refine word choices, or improve tone. But more recently, I’ve relied on it for more sophisticated tasks: summarizing articles, identifying patterns across readings, and even drafting significant portions of essays. While I haven’t used it to write a full thesis, I suspect the only reason is that the free version I use isn’t as powerful as the paid ones. I sometimes wonder whether this tool has become a crutch. Yet paradoxically, this crutch has been stabilizing. It allows me to reach “good enough” more efficiently, which frees up time and mental energy for the areas I care about most.

One of those areas is Chinese, a language I’ve studied formally since middle school and continued to pursue in college, even after fulfilling the language requirement. I’ve used AI to write Chinese essays more quickly—especially during semesters when I was juggling jobs, other coursework, and mental health struggles. Still, my commitment to learning Chinese has never wavered. I have invested not only in grammar and vocabulary, but in the culture and history embedded in the language. Passing two advanced Chinese classes at Georgetown met the requirement for my major, but it did not make me feel genuinely proficient. Perhaps that is partly due to my reliance on AI for assignments. To deepen my fluency, I enrolled in a rigorous summer study abroad program in Beijing.

That decision reflects what AI has ultimately taught me: how to clarify what I truly value. By alleviating some cognitive burdens, it has allowed me to focus more intently on passions like language learning. It has helped me distinguish between performing well and learning meaningfully.

This clarification brought to mind something I learned in a political philosophy course my freshman year. We were studying Marx, and our professor described Marx’s vision of a post-scarcity society—where technology would free humans from survival-based labor, allowing them to work only on what they love. That idea stuck with me, especially because it coincided with my first experiment using AI for academic work. That early attempt was unsuccessful—I didn’t yet know how to prompt it effectively—but the seed was planted. Could AI be one of the tools Marx imagined? One that liberates us from mechanical tasks so we can rediscover what makes us human?

This question has resurfaced across many classes I’ve taken—from Chinese discussions on regional identity to philosophy seminars on the construction of selfhood. In each context, I’ve learned that our sense of self is often defined in opposition to something else. In this way, AI has helped me better understand my own humanity. The more we use machines to handle repetitive academic tasks, the more we recognize what they cannot do: they cannot replicate intellectual curiosity, emotional insight, ethical discernment, or personal voice. Ironically, it is through contrast with AI that we begin to see which parts of learning are most human—and most irreplaceable.

I was reminded of this again recently, during a hiking trip while studying abroad in China. I found myself in conversation with a young professional who worked as a teacher. Though we didn’t know each other well and the exchange wasn’t particularly emotional, it left a lasting impression. As we walked, we discussed the relationship between education, culture, and economic pressure. We noticed a shared paradox in both the U.S. and China: students regularly use AI to complete their work, yet they also fear AI as a future job-killer. The labor market is tightening, and young people in both countries—whether in Shanghai or New York—are feeling squeezed. He mentioned that his students talk about AI with the same mix of practicality and unease that mine do. I added that in the U.S., this anxiety is further intensified by a rural–urban divide. While rural students often blame foreign competition, urban students increasingly see AI itself as the main threat to their hard-earned degrees.

That conversation didn’t yield solutions. But it reinforced my belief that the challenge ahead is not to reject AI outright. Instead, it is to use it wisely—to understand when it is helpful, when it is limiting, and when it is leading us away from what truly matters. I don’t believe we will stop using AI anytime soon. Nor am I convinced that we should. But I do believe we must continue asking the hard questions about its role in education and in our lives.

For me, AI has become a mirror. It has revealed the ways I struggle with perfectionism, highlighted my deeper commitments, and sharpened my understanding of what real learning looks like. I remain cautiously optimistic about AI’s role in education—not because it will solve all our problems, but because it challenges us to ask what education is really for. And for me, that answer has something to do with curiosity, with passion, and with the deeply human desire to understand ourselves and others, even as the world—and the tools we use to navigate it—continues to change.▪