Facilitating Learning Part 4: The AI Revolution and Why STEM Education Holds The Keys to Our Future 《促进学习》第四部分：人工智能革命以及为什么 STEM 教育是我们未来教育的关键？

In our previous articles, we have explored how AI is changing how we look at education for better or worse. We have also looked at what AI can do and what it cannot do in layman’s terms. However, one very important realization must be made before we move on! 

在前几篇文章中，我们探讨了人工智能如何改变我们对教育的看法，无论好坏。我们还通俗地探讨了人工智能能做什么和不能做什么。不过，在继续讨论之前，我们必须认识到一个非常重要的问题！

While AI can analyze data and create new content based on information within its database in near record speed, and it benefits from an interconnected network of computers, which allows it to access the sum total of human knowledge and innovation, it is still limited to what we already know. Therefore, it can generate seemingly original media across multiple mediums, with a nearly endless range of possible permutations of the data; however, it cannot wonder or think, and as a result, it cannot innovate for itself. It cannot advance human knowledge; it can only benefit from it. Therefore, careers in the fields of STEM are more important now than they have ever been! After all, critical thinking and innovation are things that we are uniquely qualified for as a species. In contrast, the fields of data and analytics have always been better suited to the unique nature of the machine. Thus, as AI continues to eat away at specific types of careers, we will also see new careers flourishing to fill in the resulting void. As a result, the educational system will need to learn to surrender the domain of data and analytics to the machine while capitalizing on and fostering the development of what computers cannot do well — Critical Thinking and innovation! 

虽然人工智能能够以近乎创纪录的速度分析数据并根据数据库中的信息创建新内容，而且它还得益于计算机互连网络，这使它能够获取人类知识和创新的总和，但它仍然局限于我们已经知道的东西。因此，它可以通过多种媒介生成看似新颖的媒体，几乎可以对数据进行无穷无尽的排列组合；但是，它不能思考，也不会思考，因此无法自我创新。它无法推动人类知识的发展，只能从中受益。因此，科学、技术、工程和数学领域的职业现在比以往任何时候都更加重要！毕竟，批判性思维和创新是我们人类独有的能力。相比之下，数据和分析领域总是更适合机器的独特性质。因此，随着人工智能不断侵蚀特定类型的职业，我们也将看到新的职业蓬勃发展，以填补由此产生的空白。因此，教育系统将需要学会把数据和分析领域拱手让给机器，同时利用和培养计算机无法胜任的领域–批判性思维和创新！

However, you might ask, “Hasn’t AI already written numerous research papers, and has it not even discovered new chemical compounds already?”. And the answer to those questions is yes, AI algorithms have done both. However, it is different from what it seems. In the case of research papers, particularly in the humanities, many papers are written solely as a secondary data source, which collates the findings from other articles and formulates an evidenced-based opinion. That being said, in these research papers, no new or original data has been generated, which AI can do quite well. In these situations, a user can develop a precise query, and the AI algorithm can then cross-reference the query against all openly accessible data sources and generate an analytical response summarizing the data in a way that supports the users’ original prompt (operator bias and all). We have even had AI algorithms lead to the discovery of new chemical compounds.

然而，你可能会问：”人工智能不是已经写出了无数的研究论文，甚至已经发现了新的化合物吗？而这些问题的答案是肯定的，人工智能算法已经做到了这两点。不过，情况与想象的有所不同。就研究论文而言，尤其是人文学科的研究论文，很多论文的撰写完全是作为第二手资料，整理其他文章的研究成果，形成有据可依的观点。也就是说，在这些研究论文中，没有产生新的或原创的数据，而人工智能可以很好地做到这一点。在这种情况下，用户可以提出一个精确的查询，然后人工智能算法可以将查询与所有公开的数据源进行交叉引用，并生成一个分析性回复，以支持用户原始提示的方式（操作员偏见等）对数据进行总结。我们甚至通过人工智能算法发现了新的化合物。

As such, you might think, what can AI not do, and is there anything left for humans to do and discover? Well, you will be happy to know there is still a need for human researchers in the fields of STEM. Not only that, the need for STEM professionals is actually more significant now than it has ever been! However, before we move on, let us explore the example of a computer AI discovering a new chemical compound in a bit more details. In this scenario, the AI proposed a new chemical compound that was theoretically possible but had yet to be discovered. However, the AI does not know what the compound could be used for, only that it was a theoretical possibility, and that is where human researchers come in. You might ask how this is even possible. How could a computer algorithm discover a new chemical element? The answer is very simple: the AI was working its way through every chemical permeation that would be possible given the list of all known elements and how they 

因此，你可能会想，人工智能还有什么不能做的，还有什么需要人类去做去发现的吗？那么，你会很高兴地知道，在 STEM 领域仍然需要人类研究人员。不仅如此，现在对 STEM 专业人才的需求实际上比以往任何时候都要大！不过，在继续讨论之前，让我们先详细探讨一下计算机人工智能发现新化合物的例子。在这种情况下，人工智能提出了一种理论上可行但尚未被发现的新化合物。然而，人工智能并不知道这种化合物可以用来做什么，只知道这是一种理论上的可能性，而这正是人类研究人员的作用所在。你可能会问，这怎么可能呢？计算机算法怎么可能发现一种新的化学元素？答案非常简单：根据所有已知元素的清单以及它们与元素周期表中其他元素的反应，人工智能正在研究每一种可能的化学渗透现象

Let me give you an example. Elements in the periodic table’s first column react with elements in the seventeenth column of the periodic table (i.e., the Halogens). For instance, Hydrogen will react with Chlorine to create hydrogen chloride (HCl) and with Florine to create Hydrogen fluoride (HF). Moreover, when both of these elements are dissolved in water, they create hydrochloric acid and hydrofluoric acid, respectively. Moving down the first column (the Alkaline Earth Metals), Lithium, which is the next element on the periodic table, will react with Chlorine to create Lithium Chloride (LiCl) or with Bromine to create Lithium Bromide (LiBr). Therefore, even if we did not know of the existence of a particular compound, we understand how the elements react with one another, and we can predict chemical compounds based on sound mathematical principles. That is how new AI algorithms were able to discover a new chemical element. 

我来举个例子。元素周期表第一列中的元素会与元素周期表第十七列中的元素（即卤素）发生反应。例如，氢会与氯反应生成氯化氢（HCl），与氟反应生成氟化氢（HF）。此外，当这两种元素溶于水时，会分别生成盐酸和氢氟酸。沿着第一列（碱土金属）往下看，周期表中的下一个元素锂会与氯反应生成氯化锂（LiCl），或与溴反应生成溴化锂（LiBr）。因此，即使我们不知道某种特定化合物的存在，我们也了解元素之间是如何反应的，我们可以根据合理的数学原理预测化合物。新的人工智能算法就是这样发现新化学元素的。

In this case, the AI did not discover a new chemical compound per se; instead, it produced a mathematical combination of theoretically possible elements. However, how did the AI know that it had discovered a new chemical compound? The answer is simple: the AI was programmed to systematically run through every chemical combination that would be possible based on our current understanding of chemistry. After identifying each new permutation, the AI would cross reference the chemical equation against all published research papers and corporate patents. If no match was found in any of the numerous global databases, the AI would flag the chemical compound for future research. However, the AI still does not know how to create this new compound, whether it is safe, highly volatile, dangerous, or even what it could be used for. That is where researchers in the fields of STEM come in. Therefore, the AI did not truly make a new discovery; it just expedited the preliminary research stage by completing tedious and mundane tasks. As such, the AI algorithm has only helped to expedite research, which will only quicken the rate of scientific advancement while also increasing the number of researchers we will need to explore all the chemical permeations that were theoretically possible but left undiscovered. It also means that researchers can focus on conducting novel research rather than cross-referencing possible chemical compounds against past discoveries to see if investigating a possible chemical formulation even warrants the investment of their time. Let us go one step further, though, as AI can only think in terms of mathematical and analytical structures and lacks creativity and the ability to think critically. 

在这个案例中，人工智能本身并没有发现一种新的化合物，而是产生了一种理论上可能存在的元素的数学组合。然而，人工智能是如何知道自己发现了一种新化合物的呢？答案很简单：根据我们目前对化学的理解，人工智能被设定为系统地运行每一种可能的化学组合。在确定每一种新的排列组合后，人工智能会将化学方程式与所有已发表的研究论文和公司专利进行交叉比对。如果在众多全球数据库中都没有找到匹配的信息，人工智能就会将该化合物标记出来，供今后研究之用。然而，人工智能仍然不知道如何制造这种新化合物，也不知道它是否安全，是否极易挥发，是否危险，甚至不知道它可以用来做什么。这就是 STEM 领域研究人员的作用所在。因此，人工智能并没有真正做出新的发现；它只是通过完成繁琐而平凡的任务，加快了初步研究阶段的进度。因此，人工智能算法只是帮助加快了研究，这只会加快科学进步的速度，同时也会增加我们需要的研究人员数量，以探索所有理论上可能存在但尚未发现的化学渗透。这也意味着研究人员可以集中精力开展新颖的研究，而不是将可能的化合物与过去的发现进行交叉对比，以确定是否有必要投入时间研究一种可能的化学配方。不过，让我们更进一步，因为人工智能只能从数学和分析结构的角度进行思考，缺乏创造力和批判性思维能力。

Perhaps one of the best-known chemical discoveries was made by Dr. Harry Coover in 1942. Invented by accident, Super Glue™ has gone on to become a household item with uses ranging from simple woodworking and appliance repair to industrial binding and medical applications. Super Glue™ was never meant to be an adhesive. During World War II, Coover was part of a team researching cyanoacrylates in an effort to find a way to make a clear plastic that could be used for precision gunsights; however, they discovered that the compound was extremely sticky, and they rejected cyanoacrylates as a feasible option before moving on with their research. Six years later, in 1951, Coover was working with a group of chemists who were researching heat-resistant polymers for jet airplane canopies. They tested cyanoacrylate monomers and realized that their unique properties required no heat or pressure to create a strong and secure bond. This kind of novel innovation comes from creative and critical thinking. An idea that an experienced researcher can formulate based on years of experience, but an AI algorithm bound by established rules and numbers would be unable to predict independently.

最著名的化学发现之一也许是哈里-库弗博士在 1942 年发现的。超级胶水™ 是一次偶然的发明，它已成为一种家喻户晓的物品，用途从简单的木工和电器维修到工业粘合和医疗应用，无所不包。超级强力胶™原本并不是一种粘合剂。在第二次世界大战期间，库弗是一个研究氰基丙烯酸酯的小组的成员，该小组试图找到一种方法来制造一种可用于精确瞄准镜的透明塑料；然而，他们发现这种化合物粘性极强，因此在继续研究之前，他们拒绝将氰基丙烯酸酯作为一种可行的选择。六年后的 1951 年，库弗与一群化学家一起研究喷气式飞机顶篷的耐热聚合物。他们测试了氰基丙烯酸酯单体，发现这种单体具有独特的性能，无需加热或加压即可形成牢固安全的粘合。这种新颖的创新来自于创造性和批判性思维。经验丰富的研究人员可以根据多年的经验提出一个想法，但受既定规则和数字束缚的人工智能算法却无法独立预测。

As we can see in all the examples in this series of short articles, AI is changing the way that we are doing things, and as a result, the way we think about education is bound to change as well. However, while these changes may seem intimidating initially, we have always had the tools we need to effectively deal with the AI revolution at our disposal. The Engineering and Design Process, the Scientific Method, and STEM education have always focused on developing and fostering inquiry and critical thinking, and that is something that AI cannot replace —. In contrast, knowledge recall and analytical thinking are the domain of the machine. There is no point for humans to try to compete on an uneven playing field where the odds are stacked against us. Instead, we need to focus on experiential learning. To capitalize on what we as humans are uniquely qualified for, and ironically, we already have the tools that we need at our disposal. We always had them; we just may not have realized it!

从本系列短文的所有例子中我们可以看到，人工智能正在改变我们做事的方式，因此，我们思考教育的方式也必将随之改变。然而，尽管这些变化最初可能看起来令人生畏，但我们一直都拥有有效应对人工智能革命所需的工具。工程与设计过程、科学方法和 STEM 教育一直注重开发和培养探究与批判性思维，这是人工智能无法替代的。相比之下，知识记忆和分析性思维则是机器的专长。人类没有必要试图在一个不公平的竞争环境中进行竞争，因为这对我们不利。相反，我们需要专注于体验式学习。利用我们人类独有的能力，而讽刺的是，我们已经拥有了我们所需的工具。我们一直都拥有这些工具，只是我们没有意识到而已！