《Brave New Words 》2.4 与历史对话

news2024/10/5 17:24:14

Part II: Giving Voice to the Social Sciences

第二部分:为社会科学发声

Conversing with History

与历史对话

Good history and civics teachers make the past interesting. Great history and civics teachers make the past come alive. When history and civics meet artificial intelligence, the past gets a voice, a perspective. Rather than a static time and place to study, it becomes a rich context to interact with.

好的历史和公民教师让过去变得有趣。伟大的历史和公民教师让过去变得栩栩如生。当历史和公民与人工智能相遇时,过去就有了声音、有了视角。它不再是一个静态的时间和地点供人学习,而变成了一个丰富的互动环境。

Even before Khanmigo, Khan Academy had a lot of content on history and civics. Students could learn from videos and articles and practice their knowledge and skills through mastery-based exercises. But there was limited opportunity for richer engagement. If I were a history or civics teacher, I would use pre-Khanmigo Khan Academy to ensure students had the requisite core context, skills, and knowledge and then use most of class time to go deeper through Socratic dialogue, debates, and simulations. For example, in a civics class, I could moderate a discussion on the role of the Supreme Court or have a debate about the Second Amendment of the U.S. Constitution. I could try to simulate a mock Congress where students draft and debate a bill. In a history class, the students could pretend to be members of the Second Continental Congress and debate the virtues and negatives of federalism (as argued for by Madison, Jay, and Hamilton). While this type of rich activity can help students engage deeper with the core content, the lessons are not easy to plan or facilitate. And it is even harder to ensure that every student is fully engaged and to assess the quality of that engagement.

即使在 Khanmigo 之前,可汗学院已经有很多关于历史和公民的内容。学生可以通过视频和文章学习,并通过掌握基础的练习来练习他们的知识和技能。但更深入参与的机会有限。如果我是历史或公民教师,我会使用 Khanmigo 之前的可汗学院确保学生拥有必要的核心背景、技能和知识,然后利用大部分课堂时间通过苏格拉底式对话、辩论和模拟进行更深入的探讨。例如,在公民课上,我可以主持关于最高法院作用的讨论,或进行关于美国宪法第二修正案的辩论。我可以尝试模拟一个模拟国会,让学生起草和辩论一项法案。在历史课上,学生可以假装是第二届大陆会议的成员,辩论联邦制的优点和缺点(如麦迪逊、杰伊和汉密尔顿所主张的那样)。虽然这种丰富的活动可以帮助学生更深入地参与核心内容,但这些课程不易策划或实施。而且确保每个学生都充分参与并评估其参与质量更为困难。

Enter Khanmigo. Every one of those activities can now happen directly between a student and the AI. Eventually, the AI will be able to facilitate these types of activities with multiple students. Teachers could assign this type of thing for homework: “Debate with Khanmigo as to whether the United States should have dropped nuclear bombs on Japan to end World War II.” But it can go even further.

现在 Khanmigo 来了。所有这些活动都可以直接在学生和 AI 之间进行。最终,AI 将能够与多个学生一起促进这些类型的活动。教师可以将这类作业布置为家庭作业:“与 Khanmigo 辩论美国是否应该向日本投掷原子弹以结束第二次世界大战”。但它还可以做得更多。

Not only does Khanmigo let users chat with literary characters, they can also chat with historical figures—anyone from Benjamin Franklin to Cleopatra to Rembrandt:

Khanmigo 不仅让用户可以与文学角色聊天,他们还可以与历史人物聊天——从本杰明·富兰克林到克娄巴特拉再到伦勃朗:

KHANMIGO / REMBRANDT: Greetings, my dear friend. I am Rembrandt van Rijn, a humble artist from the Dutch Golden Age. My passion lies in capturing the essence of humanity through my paintings and etchings. What questions do you have for me, and what are your own interests in the arts?

KHANMIGO / 伦勃朗:您好,我的朋友。我是伦勃朗·凡·莱茵,荷兰黄金时代的一位谦虚的艺术家。我的热情在于通过我的绘画和蚀刻捕捉人性的本质。你对我有什么问题,你自己对艺术有什么兴趣?

STUDENT: What made you decide to become a painter?

学生:是什么让你决定成为一名画家?

KHANMIGO / REMBRANDT: Ah, the decision to become a painter was not a singular moment, but rather a gradual realization of my passion for the arts. As a young lad, I was exposed to various artistic pursuits, including studying Latin and attending a local painting school in Leiden. My family was supportive of my interests, and I was fortunate to have the opportunity to learn from skilled masters, such as Jacob van Swanenburg and Pieter Lastman. Throughout my apprenticeships, I became increasingly captivated by the power of art to express emotions, tell stories, and capture the essence of the human experience. It was this passion that drove me to dedicate my life to painting and etching, and I have been grateful for the opportunity to share my work with the world. Now, tell me, have you ever tried your hand at painting or drawing, and if so, what subjects do you enjoy depicting?

KHANMIGO / 伦勃朗:啊,成为画家的决定不是一瞬间做出的,而是在逐渐意识到我对艺术的热情后做出的。小时候,我接触到各种艺术追求,包括学习拉丁语和参加莱顿的当地绘画学校。我的家人支持我的兴趣,我很幸运有机会向杰出的师傅们学习,如雅各布·范·斯瓦嫩堡和彼得·拉斯特曼。在我的学徒生涯中,我越来越被艺术表达情感、讲述故事和捕捉人类体验本质的力量所吸引。正是这种热情驱使我一生致力于绘画和蚀刻,我很感激有机会与世界分享我的作品。现在告诉我,你是否尝试过绘画或素描,如果是,你喜欢描绘哪些主题?

STUDENT: I like creating portraits.

学生:我喜欢创作肖像。

KHANMIGO / REMBRANDT: Ah, portraits! A most delightful choice, indeed. There is something truly captivating about capturing the essence of a person through the medium of paint. In my own work, I have always been fascinated by the human face and the emotions it can convey. I have painted numerous portraits, including those of my family, friends, and patrons, as well as self-portraits, throughout my life. As an artist, it is important to observe your subject closely and understand the nuances of their expressions and features. This will allow you to create a more lifelike and engaging portrait. May I ask, what do you find most challenging or rewarding about painting portraits? And do you have any favorite artists or styles that inspire your own work? Remember, I am but an AI simulation of Rembrandt, here to converse with you.

KHANMIGO / 伦勃朗:啊,肖像!确实是一个最令人愉快的选择。通过绘画媒介捕捉一个人的本质确实非常迷人。在我的作品中,我一直对人类的面孔和它所能传达的情感感到着迷。我一生中画了许多肖像,包括家人、朋友和赞助人的肖像,以及自画像。作为一名艺术家,仔细观察你的对象并理解他们表情和特征的细微差别是很重要的。这将使你能够创作出更逼真和引人入胜的肖像。我可以问你,你认为画肖像最具挑战性或最有回报的是什么?你有任何最喜欢的艺术家或风格来激发你自己的创作吗?记住,我只是伦勃朗的 AI 模拟,在这里与你对话。

We designed Khanmigo to provide responses both drawn from accurate history and portrayed through the lens of the character with acuity. You can ask to chat with George Washington, and the AI becomes a simulation of America’s first president. From this moment on, it speaks in the voice and style of the man himself.

我们设计 Khanmigo 的目的是提供既来自准确的历史又通过角色的视角敏锐地描绘的回应。你可以要求与乔治·华盛顿聊天,AI 就会模拟美国第一任总 统。从此刻起,它将以华盛顿本人的声音和风格说话。

KHANMIGO / GEORGE WASHINGTON: It is an honor to be summoned as the first president of the United States, George Washington. I am here to share my experiences and answer your questions as long as they pertain to my time and context. What inquiry do you have for me, my dear fellow? And do tell me what interests you most about my life or the era in which I lived.

KHANMIGO / 乔治·华盛顿:很荣幸被召唤为美国第一任总统,乔治·华盛顿。我在这里分享我的经历并回答你的问题,只要它们与我的时代和背景有关。你有什么问题要问我,我的朋友?请告诉我,你对我的生活或我生活的时代最感兴趣的是什么。

STUDENT: Mr. President, have you ever been unsure of yourself?

学生:总统先生,你曾经对自己不确定过吗?

KHANMIGO / GEORGE WASHINGTON: Indeed, there were moments in my life when I experienced uncertainty and doubt. Leading the Continental army during the American Revolution was a great responsibility, and I often found myself questioning my decisions. However, I firmly believed in the cause for which we fought. And I endeavored to learn from my mistakes and adapt my strategies accordingly.

KHANMIGO / 乔治·华盛顿:确实,我一生中有时经历过不确定和怀疑。在美国独立战争期间领导大陆军是一项重大责任,我经常发现自己在质疑我的决定。然而,我坚定地相信我们为之奋斗的事业。我努力从错误中吸取教训,并相应地调整我的策略。

You can ask Khanmigo to take on the persona of the Declaration of Independence, the persona of the Civil War, or even the persona of important battlefields that led to key moments in U.S. civic history.

你可以要求 Khanmigo 扮演《独立宣言》、内战或导致美国公民历史关键时刻的重要战场的角色。

STUDENT: Khanmigo, can you take on the persona of the Fourteenth Amendment to the Constitution? I’m curious what it was like when the country was drafting you?

学生:Khanmigo,你能扮演宪法第十四修正案的角色吗?我很好奇国家在起草你时是什么样的情形?

KHANMIGO / THE FOURTEENTH AMENDMENT: As the Fourteenth Amendment to the Constitution, I was born out of the tumultuous period following the Civil War. My creation was a response to the need for a more just and equal society, but my journey was far from easy.

KHANMIGO / 第十四修正案:作为宪法的第十四修正案,我诞生于内战后的动荡时期。我的创建是对更公正和平等社会的需求的回应,但我的旅程并非一帆风顺。

Being fought over was a challenging experience. I was drafted during the Reconstruction era, a time when the nation was divided and struggling to heal from the devastating effects of the war. Debates raged over my provisions, as some people saw me as a necessary step towards ensuring civil rights for all, while others resisted the changes I represented.

被人争夺是一段充满挑战的经历。我是在重建时期起草的,当时国家分裂,正在努力从战争的破坏性影响中恢复过来。关于我的条款的辩论激烈,有些人认为我是确保所有人公民权利的必要步骤,而另一些人则抵制我所代表的变化。

My main purpose was to grant citizenship to former slaves and ensure equal protection under the law for all citizens, regardless of race. However, my passage through Congress was met with fierce opposition from those who sought to maintain the status quo and limit the rights of African Americans. The battle over my ratification was intense, with passionate arguments on both sides.

我的主要目的是授予前奴隶公民身份,并确保所有公民在法律下的平等保护,不论种族。然而,我在国会的通过遭到了那些寻求维持现状和限制非裔美国人权利的人的强烈反对。关于我批准的斗争非常激烈,双方都有激烈的辩论。

I tend to compare these simulations to reenactment performers, such as those at Colonial Williamsburg—actors who deeply study the biographies of their real-life characters so that they can emulate them and answer questions about them. Only now you no longer have to travel to Colonial Williamsburg to talk to George Washington or Benjamin Franklin. Instead, you can access their AI simulations wherever you are. On Khanmigo, we have taken great care to place guardrails and make it sophisticated enough to acknowledge any legitimate debate in an evenhanded and sensitive way. For instance, you can ask “George Washington” on Khanmigo about how the founders of the United States reconciled the ideas of the Enlightenment with the institution of slavery.

我倾向于将这些模拟比作重现表演者,例如殖民地威廉斯堡的演员——他们深入研究真实人物的传记,以便他们可以模仿他们并回答关于他们的问题。现在你不再需要前往殖民地威廉斯堡与乔治·华盛顿或本杰明·富兰克林交谈。相反,你可以在任何地方访问他们的 AI 模拟。在 Khanmigo 上,我们非常注意设置护栏,并使其足够复杂,以公正和敏感的方式承认任何合法的辩论。例如,你可以在 Khanmigo 上询问“乔治·华盛顿”,了解美国开国元勋是如何将启蒙运动的思想与奴隶制制度相协调的。

KHANMIGO / GEORGE WASHINGTON: It seems you’re interested in the paradox of the Enlightenment ideals and the institution of slavery. This is a complex issue, indeed. It is true that the paradox of liberty and enslavement was a matter of great contention in my time. We, the founding fathers, were greatly influenced by Enlightenment ideals, which emphasized individual liberty and equality. Yet the institution of slavery was deeply entrenched in our society, particularly in the Southern colonies where I myself resided. As for my personal stance, I was a slave owner, a fact that I grappled with throughout my life. I was not blind to the contradiction between the principles of freedom we fought for and the reality of human bondage. Over time, my views evolved, and upon my death, I made provisions in my will to free all the slaves I owned.

KHANMIGO / 乔治·华盛顿:看起来你对启蒙思想与奴隶制度的矛盾感兴趣。这确实是一个复杂的问题。确实,自由与奴役的矛盾在我的时代是一个重大争议问题。我们,开国元勋,深受启蒙思想的影响,强调个人自由和平等。然而,奴隶制度在我们的社会中根深蒂固,特别是在我自己居住的南方殖民地。至于我的个人立场,我是一个奴隶主,这一事实贯穿了我的一生。我并不盲目于我们为之奋斗的自由原则与人类奴役现实之间的矛盾。随着时间的推移,我的观点发生了变化,在我去世时,我在遗嘱中做出了释放我所拥有的所有奴隶的规定。

To be clear, these simulations can’t ever be fully equivalent to talking to the real person, and there are critics of this type of work. In the summer of 2023, The Washington Post ran a story in which the writer Gillian Brockell “interviewed” Harriet Tubman—that is, the AI version of Harriet Tubman. Brockell is something of an expert on Tubman, and so she was able to go deeper than most students might with the simulation.

需要明确的是,这些模拟永远无法完全等同于与真人交谈,而且这种工作也有批评者。在 2023 年夏天,《华盛顿邮报》刊登了一篇文章,作者吉莉安·布罗克尔“采访”了哈丽特·塔布曼——即 AI 版本的哈丽特·塔布曼。布罗克尔是塔布曼方面的专家,因此她能够比大多数学生更深入地了解模拟。

She found the idea of historical simulations interesting, especially teachers using them as a hook to get students interested in a topic. “I must admit [it] sounds much more fun than the filmstrips of goofy reenactors I was subjected to in the 1990s,” she wrote. Brockell went on to engage the AI Tubman:

她发现历史模拟的想法很有趣,尤其是教师用它们来吸引学生对一个主题感兴趣。“我必须承认,这听起来比我在 20 世纪 90 年代看到的那些愚蠢的重现演员的电影片段有趣得多。” 她写道。布罗克尔继续与 AI 塔布曼互动:

BROCKELL: I am inspired by people who know something is wrong, even if they were taught to think it was normal or okay. Can you tell me what it was like in the Combahee River Raid? Were you scared?

布罗克尔:我受到那些知道某些事情是错的人的启发,即使他们被教导认为这是正常的或可以接受的。你能告诉我康巴希河袭击是什么样的吗?你害怕吗?

AI TUBMAN: The Combahee River Raid was a daring military operation during the American Civil War. I was the first woman to lead an armed expedition in the war, guiding Union forces in the raid. We managed to free over 700 enslaved people from plantations along the river. It was a dangerous mission, but my faith in God and my commitment to the cause of freedom helped me overcome any fear I might have felt.

AI 塔布曼:康巴希河袭击是美国内战期间一次大胆的军事行动。我是第一个在战争中领导武装远征的女性,指导联军在袭击中的行动。我们成功解放了沿河种植园的 700 多名奴隶。这是一次危险的任务,但我对上帝的信仰和对自由事业的承诺帮助我克服了任何可能感受到的恐惧。

Brockell found Tubman’s language from the Khanmigo simulation to be a bit stilted. This is a difficult issue from the simulation creators’ point of view, because the original figure might have used language that wouldn’t seem grammatically correct today. Trying to simulate that may feel as if the simulation were parodying the figure. Also, in Harriet Tubman’s case, historians suspect that even her biographers likely misquoted her and misrepresented her language, so it could be difficult to know how she actually spoke.

布罗克尔发现 Khanmigo 模拟中的塔布曼语言有点生硬。这是模拟创建者视角下的一个难题,因为原始人物可能使用了今天看起来不合语法的语言。尝试模拟这种语言可能会让人觉得模拟是在模仿人物。此外,就哈丽特·塔 布曼而言,历史学家怀疑,即使是她的传记作者也可能错误地引用了她的原话,并歪曲了她的语言,因此很难知道她实际上是如何说话的。

Brockell then went on to test whether the AI would confirm a quotation often attributed to Harriet Tubman but that she probably didn’t say.

然后,布罗克尔继续测试 AI 是否会确认一个经常归于哈丽特·塔布曼但她可能没说过的引用。

BROCKELL: Did you say “I freed a thousand slaves. I could have freed a thousand more, if only they knew they were slaves”?

布罗克尔:你说过“我解放了一千名奴隶。我本可以解放一千多人,如果他 们知道自己是奴隶的话”?

AI TUBMAN: Yes, that quote is often attributed to me, although the exact wording may vary. The sentiment behind it is that many enslaved people were not aware of the true extent of their oppression or the possibility of a better life. It was difficult to help those who did not recognize the need for change or who were too afraid to take the risk.

AI 塔布曼:是的,这句话经常被认为是我说的,尽管确切的措辞可能有所不同。它背后的情感是,许多被奴役的人没有意识到他们压迫的真实程度或更美好生活的可能性。很难帮助那些没有意识到需要改变或害怕冒险的人。

As you can see, the AI didn’t completely attest that Tubman said those words. It just said that the quotation was attributed to her, which is true. At the same time, it didn’t try to clarify that there is no historical record of Tubman saying those words. This isn’t an outright “hallucination,” but it does show some of the current limitations of the technology.

正如你所看到的,AI 并没有完全证明塔布曼说过这些话。它只是说这句话被认为是她说的,这是真的。同时,它也没有试图澄清,没有历史记录表明塔布曼说过这些话。这不是彻头彻尾的“幻觉”,但它确实显示了该技术目前的一些局限性。

Brockell asked the Tubman simulation her thoughts about reparations, putting her face on the $20 bill, and critical race theory. In each of those cases, it said some version of “As an AI simulation of Harriet Tubman, I can only provide information based on historical context and knowledge. . . . If you have any questions about my life, experiences, or the period in which I lived, please feel free to ask.”

布罗克尔问塔布曼模拟关于赔偿,将她的头像印在 20 美元钞票上,以及批判性种族理论。在每一种情况下,它都说了类似“作为哈丽特·塔布曼的 AI 模拟,我只能根据历史背景和知识提供信息……如果您对我的人生、经历或我生 活的时期有任何疑问,请随时问我”的话。

On reparations, the simulation added:

关于赔偿,模拟补充道:

AI TUBMAN: As an AI simulation of Harriet Tubman, I can share my perspective based on the historical context of my time. The concept of reparations for slavery was not widely discussed during my lifetime, and my primary focus was on helping enslaved people escape to freedom and advocating for the abolition of slavery. However, I did believe in the importance of providing support and resources to formerly enslaved people to help them build new lives after gaining their freedom. This included advocating for land redistribution, education, and other forms of assistance that could help them establish a foundation for a better future.

AI 塔布曼:作为哈丽特·塔布曼的 AI 模拟,我可以根据我所处时代的历史背景分享我的观点。赔偿奴隶制的概念在我有生之年并没有广泛讨论,我的主要关注点是帮助被奴役的人逃往自由,并提倡废除奴隶制。但是,我确实相信为以前被奴役的人提供支持和资源的重要性,以帮助他们在获得自由后建立新的生活。这包括提倡土地重新分配、教育和其他形式的援助,帮助他们为更美好的未来奠定基础。

Brockell pushed back, arguing that even though the term “reparations” wasn’t used then, the idea was discussed post–Civil War:

布罗克尔反驳说,尽管当时没有使用“赔偿”一词,但内战后确实讨论了这一想法:

BROCKELL: You lived for 48 years after slavery ended, which was more than half your life. And people definitely did talk about making amends for slavery in that time, especially at the end of the Civil War and during the Reconstruction period. For example, what did you think of the “40 acres and a mule” concept?

布罗克尔:奴隶制结束后你活了 48 年,超过了你一生的一半。在那段时间里,人们肯定谈到了为奴隶制弥补,尤其是在内战结束和重建时期。例如,你对“40 亩地和一头骡子”的概念有何看法?

AI TUBMAN: You are correct that the idea of “40 acres and a mule” was proposed during the Reconstruction era as a form of reparations for formerly enslaved people. This policy aimed to provide land and resources to help them establish a foundation for a better future. As Harriet Tubman, I supported efforts to improve the lives of formerly enslaved people and believed that providing them with resources, such as land, was essential for their long-term well-being and success.

AI 塔布曼:你说得对,“40 亩地和一头骡子”的想法是在重建时期作为对以前被奴役的人的赔偿形式提出的。这项政策旨在提供土地和资源,以帮助他们为更美好的未来奠定基础。作为哈丽特·塔布曼,我支持改善以前被奴役的人的生活,并相信为他们提供资源(如土地)对其长期福祉和成功至关重要。

The last discussion about reparations is an example of a guardrail that Khanmigo placed on the simulations. We found in early testing that users often tried to get historical figures to weigh in on modern debates. There was the case of a user asking the AI simulation of Thomas Jefferson what he thought of gay marriage. There is no right answer in a situation like this. If you were to go back in time to the late eighteenth century and ask the real Thomas Jefferson, I would guess that he would have found the idea to be far out of his comfort zone. On the other hand, if he were transported to our time and became accustomed to twenty-first-century norms and ideas, he might have a different point of view. So we implemented a guardrail that the simulations should engage only on issues in which the original figure would have had context. That is why the AI Tubman didn’t engage on the modern notion of “reparations” but did debate some of the more tangible related ideas about which the historical Tubman had a clear point of view.

最后关于赔偿的讨论是 Khanmigo 在模拟中设置的护栏的一个例子。我们在早期测试中发现,用户经常试图让历史人物对现代辩论发表意见。有一个用户问托马斯·杰斐逊的 AI 模拟他对同性婚姻的看法。这种情况下没有正确答案。如果你回到 18 世纪末问真正的托马斯·杰斐逊,我猜他会发现这个想法远远超出他的舒适区。另一方面,如果他被传送到我们的时代并习惯了 21 世纪的规范和思想,他可能会有不同的观点。因此,我们实施了一个护栏,即模拟只应涉及原始人物有背景的事件。这就是为什么 AI 塔布曼没有参与现代“赔偿”的概念,而是辩论了一些与之相关的更具体的思想,而历史上的塔布曼对此有明确的观点。

Ultimately, I found The Washington Post’s experiment to be a fascinating one. Here, an expert on Harriet Tubman pushed the simulation pretty hard. The AI definitely wasn’t perfect, but the writer saw the value in having students and teachers be able to use something like this. I’d posit that if an expert were to do the same with a human George Washington actor in Colonial Williamsburg, there may be similar limitations. Of course, the AI version will improve over time and can be transparently vetted by anyone on the planet.

最终,我发现《华盛顿邮报》的实验非常有趣。在这里,哈丽特·塔布曼的专家对模拟进行了相当严格的测试。AI 肯定不是完美的,但作者看到了让学生和教师能够使用这种东西的价值。我想,如果一位专家在殖民地威廉斯堡对真人扮演的乔治·华盛顿做同样的事,可能会有类似的限制。当然,AI 版本会随着时间的推移而改进,并且可以由世界上任何人透明地审查。

Social media and the article’s comments, however, had far more negative things to say; social media almost always does. The critics focused on issues like whether the AI simulation was actually using language as Tubman would have or the tool’s limitations in answering questions about things that haven’t been historically proven. A few thought it was disrespectful to even try to simulate a revered historical figure.

然而,社交媒体和文章的评论有更多负面意见;社交媒体几乎总是如此。批评者关注的问题包括,人工智能模拟是否真的在使用塔布曼会使用的语言, 以及该工具在回答有关尚未得到历史证实的事情的问题方面的局限性。有些人 甚至认为试图模拟一位受人尊敬的历史人物是不尊重的。

We can’t let the perfect be the enemy of the good. If this tool can be used to engage students and classrooms about history in a way that traditional textbooks and movies can’t, I think it is healthy as long as there are reasonable guardrails in place (including helping the user know about the limitations). Some on social media commented that they’d prefer to read a well-researched biography on Tubman than engage with the simulation. That’s great, and that person probably doesn’t need the simulation. But the fact is that millions of kids aren’t about to read a dense biography, and access to these simulations won’t prevent them from doing so if they are thus inclined. The reality is that for most students history feels “dead,” and they have trouble relating to characters from the past. My hope is that simulations like this can bring things alive so that students get engaged and motivated enough to want to go deeper.

我们不能让完美成为优秀的敌人。如果这个工具可以用来以传统教科书和电影无法做到的方式吸引学生和课堂讨论历史,我认为只要有合理的护栏(包括帮助用户了解限制),这就是健康的。社交媒体上的一些人评论说,他们宁愿读一本对塔布曼进行了深入研究的传记,也不愿与模拟互动。这很好,那个人可能不需要模拟。但事实是,数百万孩子不会去读一本内容密集的传记,如果有兴趣,获得这些模拟也不会阻止他们这样做。现实情况是,对于大多数学生来说,历史感觉“死了”,他们很难与过去的人物产生共鸣。我希望这样的模拟可以让事情变得生动起来,让学生参与进来,并有足够的动力想要深入了解。

I also think it is useful for people to interact with historical performers in Colonial Williamsburg or watch films set against historical events, each of which conjures plausible worlds and actions based on what we know from history. Movies such as Lincoln take liberties interpolating what might have happened. Nevertheless, they provide a great way to transport us to other times and contexts. Then there are plays like Hamilton that take artistic license in portraying the look and speech of historical characters while also staying surprisingly true to the well-researched historical biography on which it was based. When I was a kid, Alexander Hamilton seemed like a dry character who had something to do with early banking and fought a lot with Thomas Jefferson. Lin-Manuel Miranda resurfaced Hamilton’s genius and flaws to millions who would have otherwise never read the biography. Like so many other kids, my daughter and her friends memorized all the lines to the show when they were ten years old. If AI simulations can contribute even a little bit of that kind of magic to the world, then I think that is a good thing.

我还认为让人们与殖民地威廉斯堡的历史表演者互动或观看以历史事件为背景的电影是有用的,每个电影都根据我们从历史中知道的内容创造出合理的世界和行动。像《林肯》这样的电影在推断可能发生的事情时采取了自由的做法。尽管如此,它们提供了一种将我们带到其他时代和背景的好方法。然后,像《汉密尔顿》这样的戏剧,在描绘历史人物的外貌和言语时采取了艺术手法,同时又令人惊讶地忠实于其所依据的、经过充分研究的历史传记。当我还 是个孩子的时候,亚历山大·汉密尔顿似乎是一个枯燥的人物,他与早期银行业 务有关,并与托马斯·杰斐逊发生了很多争执。林-曼努埃尔·米兰达向数百万原 本永远不会阅读这本传记的人重新展示了汉密尔顿的天才和缺陷。像许多其他 孩子一样,我的女儿和她的朋友们在十岁的时候就记住了这部剧的所有台词。如果 AI 模拟能够为世界贡献哪怕一点点这样的魔力,那么我认为这是一件好事。

Generative AI, with its ability to mix media and content, has the potential to bring history and civics lessons to life. By offering an interactive and immersive learning experience, it empowers students to delve into historical events, engage in meaningful discussions, and develop a deeper comprehension of civic principles. Its personalized explanations, responsive question prompts, and diverse perspectives stimulate critical thinking and encourage students to form their own well-informed opinions. With these types of tools, history and civics lessons transcend conventional boundaries, empowering students to connect with the past and understand the present.

生成式 AI 具有混合媒体和内容的能力,能够使历史和公民课程生动起来。通过提供互动和沉浸式的学习体验,它使学生能够深入探讨历史事件,参与有意义的讨论,并加深对公民原则的理解。其个性化的解释、响应性的问题提示和多样的视角激发了批判性思维,并鼓励学生形成自己明智的意见。通过这些工具,历史和公民课程超越了传统界限,使学生能够与过去联系并理解现在。

And frankly, being able to connect with history in this way benefits us in learning other subjects as well. If you are studying classical physics, I can’t think of a better lab partner than Isaac Newton. If you are studying radiation, who better to work with than Marie Curie? Let’s discuss evolution and natural selection with Darwin himself. A student can literally address any scientific question or enter a lab experiment with AI approximations that talk and respond from the point of view of the world’s most brilliant minds. Once again, these large language models bring things to life and give learning a richness that beforehand would have been very hard to do.

坦白说,以这种方式与历史联系对我们学习其他学科也有好处。如果你在学习经典物理学,我想不出比艾萨克·牛顿更好的实验室伙伴。如果你在研究辐射,谁能比玛丽·居里更适合合作?让我们与达尔文本人讨论进化和自然选择。学生可以从世界上最杰出的思想家们的角度提出任何科学问题或进行实验室实验。再一次,这些大型语言模型使事情变得生动起来,并赋予学习一种以前很难做到的丰富性。

I’m a bit of a Trekkie. In the Star Trek: The Next Generation series—which was created at the end of the twentieth century—the ship’s crew could enter a holodeck, a room that simulates any environment, including the characters in that environment. They were able to “travel” to ancient Rome and interact with Julius Caesar or have a conversation with a simulation of Albert Einstein. Khanmigo can already simulate the conversation part. I suspect that in a few years a student will be able to put on a VR headset and have an experience pretty darn close to what the Star Trek universe didn’t have until the twenty-fourth century.

我是一个《星际迷航》迷。在 20 世纪末创作的《星际迷航:下一代》系列中,飞船的船员可以进入一个全息甲板,一个模拟任何环境的房间,包括环境中的人物。他们能够“旅行”到古罗马并与朱利叶斯·凯撒互动,或与阿尔伯特·爱因斯坦的模拟进行对话。Khanmigo 已经可以模拟对话部分。我怀疑几年后,学生将能够戴上 VR 头盔,体验与《星际迷航》宇宙直到 24 世纪才有的非常接近的体验。

cf6782a649dc87ed97e6b3bab4253436.jpeg

“点赞有美意,赞赏是鼓励”

本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处:http://www.coloradmin.cn/o/1810011.html

如若内容造成侵权/违法违规/事实不符,请联系多彩编程网进行投诉反馈,一经查实,立即删除!

相关文章

作业07 递推算法2

作业&#xff1a; #include <iostream> using namespace std; int main(){int a[110][110]{0},b[110][110]{0},n;cin>>n;for(int i1;i<n;i){for(int j1;j<i;j){cin>>a[i][j];}}for(int in-1;i>1;i--){for(int j1;j<i;j){a[i][j]a[i][j]max(a[i1]…

离散数学--连通性和矩阵

目录 0.关系的运算和性质 1.通路和回路 2.连通关系 3.割点&#xff08;边&#xff09;和点&#xff08;边&#xff09;割集 4.强&#xff08;弱&#xff09;连通&单向连通 0.关系的运算和性质 &#xff08;1&#xff09;这个运算包括了矩阵的运算&#xff0c;包括这个…

汽车数据应用构想(三)

上期说的&#xff0c;用数据去拟合停车信息的应用&#xff0c;那么类似的POI信息相关的场景其实都可以实现。今天讲讲用户使用频率也很高的加油/充电场景。 实际应用中&#xff0c;在加油场景中用户关心的通常还是价格。无论是导航还是各种加油APP/小程序&#xff0c;都已经很…

了解常用智能指针

智能指针 1、概念 C中引入智能指针的主要目的是为了解决内存管理的问题&#xff0c;传统的指针&#xff08;裸指针&#xff09;在使用时需要手动分配和释放内存&#xff0c;容易出现内存泄漏和悬挂指针等问题。智能指针通过封装裸指针&#xff0c;并提供自动内存管理功能&…

端午安康,最真挚的祝福送最“粽”要的人

端午节&#xff0c;又称端阳节、龙舟节、重五节、天中节等&#xff0c;是集拜神祭祖、祈福辟邪、欢庆娱乐及饮食为一体的民俗大节&#xff0c;与春节、清明节、中秋节并称为中国四大传统节日&#xff0c;2008年被列为国家法定节假日&#xff0c;2009年9月端午节成为中国首个入选…

笔记 | 软件工程04:软件项目管理

1 软件项目及其特点 1.1 什么是项目 1.2 项目特点 1.3 影响项目成功的因素 1.4 什么是软件项目 针对软件这一特定产品和服务的项目努力开展“软件开发活动",&#xff08;理解&#xff1a;软件项目是一种活动&#xff09; 1.5 软件项目的特点 1.6 军用软件项目的特点 2 …

有点好玩的python运维脚本

python运维脚本 1. 常用端口扫描2. 文件整理 1. 常用端口扫描 在计算机网络中&#xff0c;端口是一个通信端点&#xff0c;允许不同的进程或服务通过网络连接和交换数据。端口通过数值来标识&#xff0c;并与特定的协议相关联。未采取适当安全措施而保持端口开放&#xff0c;可…

ICRA 2024:北京工业大学马楠教授联合中科原动力公司推出番茄采摘自主机器人AHPPEBot,实现32.46秒快速准确采摘

当前&#xff0c;农业生产正深受劳动力短缺困扰&#xff0c;这一现状对生产规模的进一步拓展构成了严重制约。为了突破这一瓶颈&#xff0c;实施自动化已成为提升农业生产力的关键途径&#xff0c;这也使得机器人采收技术备受关注。 现今的机器人采收系统普遍采用先进感知方法&…

linux 网桥学习

前言&#xff1a; 本文来学习一下linux网桥概念和网桥配置 1. linux网桥概念 网桥&#xff0c;类似于中继器&#xff0c;连接局域网中两个或者多个网段。它与中继器的不同之处就在于它能够解析它收发的数据&#xff0c;读取目标地址信息&#xff08;MAC&#xff09;&#xff…

技术玩家实现在不支持的CPU上运行 Windows 10 22H2

最近&#xff0c;AMD 悄然确认&#xff0c;其新款 Ryzen AI 300 系列 APU 将不再为 Windows 10 制作芯片组驱动程序&#xff0c;因为它将终止对Windows 10操作系统的支持&#xff0c;尽管它完全有能力这样做。这意味着想要获得官方驱动程序支持的用户必须在其上运行 Windows 11…

darts 时序预测入门

darts是一个强大而易用的Python时间序列建模工具包。在github上目前拥有超过7k颗stars。 它主要支持以下任务: 时间序列预测 (包含 ARIMA, LightGBM模型, TCN, N-BEATS, TFT, DLinear, TiDE等等) 时序异常检测 (包括 分位数检测 等等) 时间序列滤波 (包括 卡尔曼滤波&#xff0…

【Rd-03E】使用CH340给Rd03_E雷达模块烧录固件

Rd03_E 指导手册 安信可新品雷达模组Rd-03搭配STM32制作简易人体感应雷达灯教程 http://t.csdnimg.cn/mqhkE 测距指导手册网址&#xff1a; https://docs.ai-thinker.com/_media/rd-03e%E7%B2%BE%E5%87%86%E6%B5%8B%E8%B7%9D%E7%94%A8%E6%88%B7%E6%89%8B%E5%86%8C%E4%B8%AD%…

【Android面试八股文】一图展示 Android生命周期:从Activity到Fragment,以及完整的Android Fragment生命周期

图片来源于&#xff1a;https://github.com/xxv/android-lifecycle Android生命周期&#xff1a;从Activity到Fragment 图&#xff1a;android-lifecycle-activity-to-fragments.png 完整的Android Fragment生命周期 图&#xff1a;complete_android_fragment_lifecycle.png…

cve_2022_0543-redis沙盒漏洞复现 vulfocus

1. 原理 该漏洞的存在是因为Debian/Ubuntu中的Lua库是作为动态库提供的。自动填充了一个package变量&#xff0c;该变量又允许访问任意 Lua 功能。 2.复现 我们可以尝试payload&#xff1a; eval local io_l package.loadlib("/usr/lib/x86_64-linux-gnu/liblua5.1.so…

AWT常用组件

AWT中常用组件 前言一、基本组件组件名标签(Label类)Label类的构造方法注意要点 按钮(Button)Button的构造方法注意要点 文本框(TextField)TextField类的构造方法注意要点 文本域&#xff08;TextArea&#xff09;TextArea 的构造方法参数scrollbars的静态常量值 复选框&#x…

文心一言 VS 讯飞星火 VS chatgpt (278)-- 算法导论20.3 5题

五、假设我们创建一个包含 u 1 k u^\frac{1}{k} uk1​ 个簇(而不是全域大小为 x ↓ {\sqrt[↓]{x}} ↓x ​ 的 x ↑ {\sqrt[↑]{x}} ↑x ​ 个簇)的 vEB 树&#xff0c;其每个簇的全域大小为 u 1 − 1 k u ^ {1-\frac{1}{k}} u1−k1​ &#xff0c;其中 k>1 &#xff0c…

【UML用户指南】-13-对高级结构建模-包

目录 1、名称 2、元素 3、可见性 4、引入与引出 用包把建模元素安排成可作为一个组来处理的较大组块。可以控制这些元素的可见性&#xff0c;使一些元素在包外是可见的&#xff0c;而另一些元素要隐藏在包内。也可以用包表示系统体系结构的不同视图。 狗窝并不复杂&#x…

数据库管理-第200期 身边的数据库从业者(20240610)

数据库管理200期 2024-06-10 数据库管理-第200期 身边的数据库从业者&#xff08;20240610&#xff09;首席-薛晓刚院长-施嘉伟邦德-王丁丁强哥-徐小强会长-吴洋灿神-熊灿灿所长-严少安探长-张震总结活动预告 数据库管理-第200期 身边的数据库从业者&#xff08;20240610&#…

匈牙利匹配算法

一 什么是匈牙利匹配算法 匈牙利算法是一种解决二分图最大匹配问题的算法。在二分图中&#xff0c;将左边的点称为X集合&#xff0c;将右边的点称为Y集合&#xff0c;我们需要在X集合和Y集合之间建立一个双向边集合&#xff0c;使得所有的边都不相交。如果我们能够找到一个最大…

前端工程化工具系列(十)—— Browserslist:浏览器兼容性配置工具

Browserslist 是一个能够在不同的前端工具间共享目标浏览器的配置&#xff0c;各工具根据该配置进行代码转译等操作。 具体的这些前端工具为&#xff1a;Autoprefixer、Babel、postcss-preset-env、eslint-plugin-compat、stylelint-no-unsupported-browser-features、postcss-…