The practical guide to using AI to do stuff 使用 AI 做事的实用指南

2023-02-23 22:37 作者:大仝tong和朋友们 0人读过 | 我要投稿

A resource for students in my classes (and other interested people).
为我班上的学生（和其他感兴趣的人）提供的资源。

The practical guide to using AI to do stuff --- 使用 AI 做事的实用指南 (substack.com)

My classes now require AI (and if I didn’t require AI use, it wouldn’t matter, everyone is using AI anyway). But how can students use AI well? Here is a basic tutorial and guide I am providing my classes. It covers some of the many ways to use AI to be more productive, creative, and successful, using the technology available in early 2023, as well as some of the risks.
我的课程现在需要 AI（如果我不需要使用 AI，那也没关系，反正每个人都在使用 AI）。但是学生如何用好AI呢？这是我提供课程的基本教程和指南。它涵盖了使用 AI 提高生产力、创造力和成功的许多方法中的一部分，使用 2023 年初可用的技术，以及一些风险。

Write stuff 写东西

Open Source Option: Nothing very good
开源选项：没什么好
Best free (for now) option: ChatGPT (registration may require a phone number)
最好的免费（目前）选项：ChatGPT（注册可能需要电话号码）
Best option if ChatGPT is down: OpenAI Playground
ChatGPT 关闭时的最佳选择：OpenAI Playground

For right now, no other general AI tool comes even close to ChatGPT, which uses a variant of the GPT-3.5 language model. You will have to register for the site, and it can be slow (especially as there is a paid tier that will get priority over free use). If ChatGPT is down, you can also access roughly same model via the OpenAI Playground, but it will not have the useful chat features. More alternatives are coming, but right now there is nothing close to the GPT models if you want to experience the power of AI.
目前，没有其他通用人工智能工具能与 ChatGPT 相提并论，后者使用 GPT-3.5 语言模型的变体。您将必须注册该网站，而且速度可能会很慢（尤其是因为有一个付费层将优先于免费使用）。如果 ChatGPT 关闭，您也可以通过 OpenAI Playground 访问大致相同的模型，但它不会有有用的聊天功能。更多的替代方案即将到来，但如果您想体验人工智能的力量，目前没有什么比 GPT 模型更接近的了。

I have experimented a lot on how to use ChatGPT to help with written material, so here is a list of way you might find it useful:
关于如何使用 ChatGPT 来帮助处理书面材料，我已经进行了很多实验，因此这里列出了您可能会发现它有用的方法：

Writing anything. Blog posts, essays, promotional material, speeches, lectures, chose-you-own adventures, scripts, short stories - you name it, it does it. But you can’t just give it basic prompts. Getting good writing out of ChatGPT takes some practice, and here is a guide to doing that. Again, though, be aware of both the ethics of passing off the work of others as your own and academic honor codes. In my class, you merely need to include a paragraph about how you used ChatGPT in any assignment, other classes may ban or restrict the tool.
写任何东西。博客文章、散文、宣传材料、演讲、讲座、自选冒险经历、剧本、短篇小说——应有尽有。但你不能只给它基本的提示。从 ChatGPT 中获得良好的写作需要一些练习，这里有一个指南。不过，请再次注意将他人的作品冒充自己的道德规范和学术荣誉准则。在我的课上，你只需要在任何作业中包括一段关于你如何使用 ChatGPT 的段落，其他班级可能会禁止或限制该工具。
Make your writing better. Paste your text into ChatGPT. Ask it to improve the content or for suggestions about how to make it better for a particular audience (though note it only “remembers” a couple thousand words of text)
让你的写作更好。将您的文本粘贴到 ChatGPT 中。要求它改进内容或就如何让它更好地为特定受众提供建议（尽管请注意它只“记住”几千字的文本）
Help you with tasks. AI can write emails, create sales templates, give you next steps in a business plan, and a lot more. Here is a guide to how to use it help in entrepreneurship that introduces some of these approaches.
帮你完成任务。 AI 可以编写电子邮件、创建销售模板、为您提供商业计划中的后续步骤等等。这是一份关于如何使用它帮助创业的指南，其中介绍了其中一些方法。
Unblock yourself. Neil Gaiman wrote: “Being a writer is a very peculiar sort of a job: it's always you versus a blank sheet of paper (or a blank screen) and quite often the blank piece of paper wins.” And I often find that this is true. Creating is hard, and having the AI suggest a header or paragraph is often a great way to move forward, even if you never use any of the text.
解锁自己。尼尔盖曼写道：“作为一名作家是一种非常特殊的工作：它总是你与一张白纸（或一个空白屏幕）的对比，而且往往是白纸获胜。”我经常发现这是真的。创建很难，让 AI 建议标题或段落通常是向前推进的好方法，即使您从不使用任何文本。

Some things to worry about: In a bid to respond to your answers, it is very easy for the AI to “hallucinate” and generate plausible facts. It can generate entirely false content that is utterly convincing. Let me emphasize that: AI lies continuously and well. Every fact or piece of information it tells you may be incorrect. You will need to check it all. Particularly dangerous is asking it for math, references, quotes, citations, and information for the internet. You are responsible for the accuracy of everything you do, don’t expect the AI to produce truthful information.
一些需要担心的事情：为了回应你的答案，人工智能很容易“产生幻觉”并产生似是而非的事实。它可以生成完全令人信服的完全虚假的内容。我要强调的是：AI 一直在撒谎，而且很好。它告诉您的每一个事实或信息都可能是不正确的。您将需要检查所有内容。特别危险的是向它索要数学、参考资料、引文、引文和互联网信息。你对你所做的一切的准确性负责，不要指望人工智能产生真实的信息。

The AI also doesn’t explain itself, it only makes you think it does. If you ask it to explain why it wrote something, it will give you a plausible answer that is completely made up. It is not interrogating its own actions, it is just generating text that sounds like it is doing so. This makes understanding biases in the system very challenging, even though those biases almost certainly exist.
人工智能也不会自我解释，它只会让你认为它会解释。如果你要求它解释为什么它写了一些东西，它会给你一个完全编造的似是而非的答案。它不是在询问自己的行为，它只是生成听起来像是在这样做的文本。这使得理解系统中的偏见非常具有挑战性，即使这些偏见几乎肯定存在。

It also can be used unethically to manipulate or cheat. You are responsible for the output of these tools.
它还可以不道德地用于操纵或欺骗。您对这些工具的输出负责。

Make images 制作图像

Open Source Option: Download Stable Diffusion
开源选项：下载 Stable Diffusion
Free option that requires sign-up: Here for DALL-E (50 images free). Here for a hosted version of Stable Diffusion that gives you 100 images.
需要注册的免费选项：这里是 DALL-E（50 张免费图片）。这里有一个托管版本的 Stable Diffusion，它可以为您提供 100 张图像。
Best option: Midjourney (25 images free, than pay)
最佳选择：Midjourney（25 张图片免费，而非付费）

There are three big image generators out there:
那里有三个大型图像生成器：

Stable Diffusion, which is open source and you can run from any high-end computer. It takes effort to get started, since you have to learn to craft prompts properly, but once you do it can produce great results. It is especially good for combining AI with images from other sources. Here is a nice guide to Stable Diffusion if you go that route (be sure to read both parts 1 and part 2).
Stable Diffusion，它是开源的，您可以在任何高端计算机上运行。开始需要付出努力，因为您必须学会正确地制作提示，但是一旦您做到了，就会产生很好的结果。它特别适合将 AI 与其他来源的图像相结合。如果你走那条路，这里有一个很好的稳定扩散指南（一定要阅读第 1 部分和第 2 部分）。
DALL-E, from OpenAI, which is a little stale but will probably be updated soon.
DALL-E，来自 OpenAI，有点陈旧，但可能很快就会更新。
Midjourney, which is the best system in early 2023. The reason I would suggest MidJourney is that it has the lowest learning-curve of any system: just type in "thing-you-want-to-see --v 4" (the --v 4 at the end is important, it uses the latest model) and you get a great result. Midjourney requires Discord. Here is a guide to using Discord. And here is a Twitter thread on using MidJourney’s interface.
Midjourney，这是 2023 年初最好的系统。我建议 MidJourney 的原因是它具有所有系统中最低的学习曲线：只需输入“thing-you-want-to-see --v 4”（ --v 4 最后很重要，它使用最新的模型），你会得到很好的结果。中途需要 Discord。这是使用 Discord 的指南。这是一个关于使用 MidJourney 界面的 Twitter 线程。

Some class-related uses: 一些类相关的用途：

Illustrate presentations or reports
说明演示文稿或报告
Create stock photographs of your product in use
创建您正在使用的产品的库存照片
Create mockups of designs 创建设计模型
Create mockups of apps or websites
创建应用程序或网站的模型
Create logos 创建徽标
Fun

Some things to worry about: These systems are built around models that have built-in biases due to their training on Internet data (if you ask it to create a picture of an entrepreneur, for example, you will likely see more pictures featuring men than women, unless you specify “female entrepreneur”), you can use this explorer to see these biases at work.
一些需要担心的事情：这些系统是围绕模型构建的，这些模型由于对互联网数据的训练而具有内在的偏见（例如，如果你要求它创建一张企业家的照片，你可能会看到更多男性图片而不是男性图片）女性，除非您指定“女性企业家”），您可以使用这个探索器来了解工作中的这些偏见。

These systems are also trained on existing art on the internet in ways that are not transparent and potentially legally and ethically questionable. Though technically you own copyright of the images created, legal rules are still hazy.
这些系统还以不透明且可能在法律和道德上有问题的方式接受互联网上现有艺术的培训。尽管从技术上讲您拥有所创建图像的版权，但法律规则仍然模糊不清。

Also, right now, they don’t create text, just a bunch of stuff that looks like text. And AI has trouble with hands.
此外，现在，他们不创建文本，只是一堆看起来像文本的东西。 AI 的手也有问题。

Come up with ideas 想出点子

Despite of (or in fact, because of) all its constraints and weirdness, AI is perfect for idea generation. You often need to have a lot of ideas to have good ideas. Not everyone is good at generating lots of ideas, but AI is very good at volume. Will all these ideas be good or even sane? Of course not. But they can spark further thinking on your part.
尽管有（或者事实上，因为）它的所有限制和怪异，人工智能非常适合创意产生。您通常需要有很多想法才能有好主意。不是每个人都擅长产生大量想法，但 AI 非常擅长数量。所有这些想法都是好的，甚至是理智的吗？当然不是。但它们可以激发您的进一步思考。

How do you do it? Just ask for ideas, and it will provide them. But there are also more advanced techniques. Here is a guide to them.
你怎么做呢？只要提出想法，它就会提供。但也有更先进的技术。这是他们的指南。

Make videos 制作视频

There are many choices, but an option I like: D-iD (note, you can create a couple videos for free with registration)
有很多选择，但我喜欢一个选项：D-iD（注意，您可以通过注册免费创建几个视频）

It is now trivial to generate a video with a completely AI generated character (you can use the images generated using the techniques in the guide), reading a completely AI-written script, talking in an AI-made voice, animated by AI. I created this video of a cyborg giving a TED Talk in under two minutes. The easiest way is to use D-iD, but a lot of competition is entering the space, and things are changing by the minute.
现在，生成具有完全由 AI 生成的角色的视频（您可以使用使用指南中的技术生成的图像）、阅读完全由 AI 编写的脚本、用 AI 制作的声音说话、由 AI 制作动画已经变得轻而易举。我制作了这个电子人在两分钟内进行 TED 演讲的视频。最简单的方法是使用 D-iD，但很多竞争正在进入这个领域，而且每时每刻都在发生变化。

This can be great for explainer videos and introductions. And within a few months you are likely to be able to generate videos from text prompts, so stay tuned.
这对于解释视频和介绍非常有用。并且在几个月内您很可能能够根据文本提示生成视频，敬请期待。

Some things to worry about: Deep fakes are a huge concern, and these systems need to be used ethically.
一些需要担心的事情：Deep fakes 是一个巨大的问题，这些系统需要以合乎道德的方式使用。

A note on coding 关于编码的说明

Both ChatGPT and GPT-3.5 are very good at writing code. There is evidence that using AI assistance in coding may cut programming time in half. You also might be able to use it even if you don’t code yourself. For example, you can create javascript art projects just by asking, as this thread explains. But the process is not 100% error-free for non-programmers.
ChatGPT 和 GPT-3.5 都非常擅长编写代码。有证据表明，在编码中使用 AI 辅助可以将编程时间缩短一半。即使您不自己编写代码，您也可以使用它。例如，您可以通过询问来创建 javascript 艺术项目，正如该线程所解释的那样。但是对于非程序员来说，这个过程并不是 100% 没有错误的。

To accomplish the same goals, non-programmers may not be aware that there has been a revolution in workable low-code or no-code application builders (many of which have nothing to do with AI) that use platforms like Bubble or Shopify to help you create fairly sophisticated apps without the need to program directly. Early evidence shows that companies built with these approaches get similar kinds of outcomes as other startups, but they do it faster and raise less cash (and get lower valuations). Many of these systems include good tutorials, such as Bubble’s Academy. However, these platforms typically charge for services.
为了实现相同的目标，非程序员可能没有意识到可用的低代码或无代码应用程序构建器（其中许多与 AI 无关）已经发生了一场革命，这些应用程序构建器使用 Bubble 或 Shopify 等平台来提供帮助您无需直接编程即可创建相当复杂的应用程序。早期证据表明，采用这些方法建立的公司获得了与其他初创公司类似的结果，但他们做得更快，筹集的现金更少（并获得更低的估值）。其中许多系统都包含很好的教程，例如 Bubble’s Academy。但是，这些平台通常对服务收费。

Learn stuff 学东西

AI can be a powerful tool for learning and exploration. I have written about how it can be used for teaching, but it can also work for self-guided learning as well. Some ways to do that in ChatGPT:
人工智能可以成为学习和探索的强大工具。我已经写过它如何用于教学，但它也可以用于自学。在 ChatGPT 中执行此操作的一些方法：

Summarize texts. I have pasted in numerous complex academic articles and asked it to summarize the results, and it does a good job! (though remember the size limits). Even better, you can then interrogate the material by asking follow-up questions: what is the evidence for that approach? What do the authors conclude? And so on…
总结课文。我已经粘贴了很多复杂的学术文章并要求它总结结果，它做得很好！（尽管记住大小限制）。更好的是，您可以通过提出后续问题来审视材料：该方法的证据是什么？作者的结论是什么？等等…
Help with concepts. You can ask the AI to explain concepts. Because we know the AI could be hallucinating, you would be wise to (carefully!) double-check its results against another source. This both helps you learn and confirms the AI output looks good. Once you have a sense that it is right, ask it to explain it in different ways “Like I am 10” or “in a script from The Office” or “in the context of a medical examination.” Again, this is a start for your learning journey, since it will often get subtleties wrong.
帮助概念。你可以要求 AI 解释概念。因为我们知道 AI 可能会产生幻觉，所以你最好（小心！）根据另一个来源仔细检查它的结果。这既可以帮助您学习，也可以确认 AI 输出看起来不错。一旦你觉得它是对的，请让它以不同的方式解释它“就像我 10 岁”或“在办公室的剧本中”或“在体检的背景下”。同样，这是您学习之旅的开始，因为它经常会出现细微的错误。
Explain errors: If you get something wrong, paste in the incorrect text and ask it to explain the error, or to walk you through it step-by-step. The results can be surprisingly helpful.
解释错误：如果你弄错了，粘贴不正确的文本并要求它解释错误，或者一步一步地引导你。结果可能非常有用。

Some things to worry about: If you don’t check for hallucinations, it is possible that you could be taught something inaccurate. Use the AI as a jumping-off point for your own research, not as the final authority on anything. Also, it isn’t actively connected to the internet, so it has no up-to-date information.
需要担心的一些事情：如果您不检查幻觉，您可能会学到一些不准确的东西。将 AI 作为你自己研究的起点，而不是任何事情的最终权威。此外，它没有主动连接到互联网，因此没有最新信息。

And more? 和更多？

This is just the beginning. AI is improving rapidly, and people are developing new uses all the time. But remember two key points:
这仅仅是个开始。人工智能正在迅速进步，人们一直在开发新的用途。但请记住两个关键点：

AI is a tool. It is not always the right tool. Consider carefully whether, given its weaknesses, it is right for the purpose to which you are planning to apply it.
人工智能是一种工具。它并不总是正确的工具。考虑到它的弱点，请仔细考虑它是否适合您计划应用它的目的。
There are many ethical concerns you need to be aware of. AI can be used to infringe on copyright, or to cheat, or to steal the work of others, or to manipulate. And how a particular AI model is built and who benefits from its use are often complex issues, and not particularly clear at this stage. Ultimately, you are responsible for using these tools in an ethical manner. Be transparent about how you use AI, and take responsibility for the output you create.
您需要注意许多道德问题。 AI 可用于侵犯版权、欺骗、窃取他人的作品或进行操纵。而一个特定的人工智能模型是如何构建的，以及谁从中受益往往是复杂的问题，在现阶段还不是特别清楚。最终，您有责任以合乎道德的方式使用这些工具。对您如何使用 AI 保持透明，并对您创建的输出负责。

We are in the early days of a very rapidly advancing revolution. Are there other uses you want to share? Let me know in the comments.
我们正处于一场快速推进的革命的初期。您还有其他用途要分享吗？在评论中让我知道。

This post is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
这篇文章是根据 Creative Commons Attribution-NonCommercial 4.0 International License 获得许可的。

标签：