Which is the best AI tool for writing? More than just one. In this overview, you’ll find the right candidates for your toolbox.

Artificial intelligence helps you write. There are many tools out there that make you write faster and better.

My team and I have tried many of them. We’ve been testing for a year. That’s why people often ask me: Which tool is the best?

Honestly, I can’t say for sure. I need to know what you want to achieve, so I can help you choose. And I know in advance that one tool won’t do it. Neither for us—nor for you.

But maybe it helps you to get a first impression if you know what I use? That’s why I’m introducing 25 tools I work with. And they’ll help you write and publish more, faster, and better.

If you’re looking for what AI tools to use to get ahead in your writing, this post will give you:

Not just one single answer. But a series of answers. There is no one tool that can do it all. That’s why we use a stack of tools in the team.

Very subjective answers. We went on a discovery journey and tested out many things. We are constantly finding new, exciting possibilities. But we haven’t reached the end yet. Everything can already change again tomorrow. And the journey begins anew.

You are looking at a snapshot of our quest, as of September 2022, which we update from time to time. It will always remain preliminary, but is still valuable. If you’re looking into how artificial intelligence can help you write for the first time, here are ideas for your own experiments.

This is an overview. It is definitely not complete, and it remains superficial in many places, while in other areas, I will go into depth. Additionally, I will explore some of the topics in detail in separate posts. That’s why we are adding more links to these deep-dives as time goes on.

What can you expect from artificial intelligence in writing?

At first, we need to take a closer look at the promise I made in the first sentence of this article. Artificial intelligence helps you write. It needs to be said explicitly: Helps you. Not: writes for you.

Many of the tools you’ll learn about in this post can write. You just need to provide some input, then they can take over and generate text. For example, turn a headline into a paragraph, rephrase an existing paragraph or summarize a complicated paragraph in a simpler way.

They produce correct grammar with believable vocabulary in the process. Generating sentence after sentence whenever you want, on and on. And they do it in a style you can actually flaunt.

However, this does not mean that you will reach your intended goal with your text. Just as the autopilot in the Tesla can drive straight ahead on the highway, keep the distance to the car in front, and even change lanes and overtake—at some point you have to steer again yourself. No later than when the curve becomes too tight on the exit ramp. That’s when the Tesla forces you to take over the wheel. At least in Europe, that’s what the law requires.

I started in late summer 2021 with the resolution to make myself obsolete. The copywriter who replaces himself—I wrote that as a mission on my website and in my social media profiles.

I didn’t succeed.

Thankfully, I’m not alone in this. The truly self-driving car is also still a pipe dream—no matter how much Elon Musk might have wished otherwise. I still do a lot of writing myself, but I’m getting help with that. And that works quite well.

When do you have to take the wheel? Before writing or after writing? What tools are there for the periods when you have to write yourself? I’ve organized this post according to these questions. That’s how these four chapters took shape:

  1. Autopilots: Let Large Language Models write for you
  2. Assistance systems: Write better, faster and more confident on your own
  3. Navigation systems: Know what’s worth writing
  4. Gigafactories: Generate thousands of texts from data

Take twenty minutes, buckle up—and let’s ride.

1. Autopilots: Let Large Language Models write for you

In the first chapter of this overview, we’ll look at tools that are based on Large Language Models. They can provide you with material for whatever you want to write. No matter what you write: seminar papers for university, a business plan for your startup, descriptive texts for your analytical report. Or, like me most of the time, marketing texts or blog posts. Set the next goal—and leave the writing to the autopilot.

Jasper: A lot of inspiration with the pioneer

Jasper is the tool I started my journey with. Jasper had a different name last year. When I launched in June 2021, it was still called Jarvis. Since then, it has evolved. But at its core, it has stayed true to itself: It’s a web-based user interface for a purchased language model. Under the hood, Jasper uses GPT-3, a Large Language Model from the provider Open AI.

Jasper is more than just software. Jasper runs a community with currently over 65,000 members on Facebook. CEO Dave Rogenmoser and CMO Austin Distel regularly produce a ton of content. Use case after use case, Jasper’s content has helped me figure out what I can do with artificial intelligence.

Jasper’s Youtube channel now features over 100 videos. A huge part of them are tutorials—but also many sessions with guests like Rachel Pedersen, the self-proclaimed Queen of Social Media.

The tool with the largest community. The tool with regularly a lot of education. The tool with inspiration for what you could do: On all three counts, the answer is always Jasper.

That says nothing about the AI as such. The content is just the framework. But a framework is worth a hell of a lot in a field where everyone is taking their first steps, and there are no standard procedures yet. Among the blind, the one-eyed is king.

Guidance, inspiration, recipes: Jasper makes it easy to tap into writing with artificial intelligence. And for me, that’s what made Jasper the tool I’ve been using the most since last summer. Jasper is versatile. But: not exactly cheap. (We’re paying $119 monthly as early adopters for three licenses and unlimited output. This unlimited plan doesn’t exist anymore for new customers; now subscriptions are tiered by volume).

Large Language Models: Many models, many resellers

GPT-3 from Open AI is not just the technical heart of the offering for Jasper—many other resellers also rely on the same engine. And GPT-3 is not the only Large Language Model. GPT stands for Generative Pretrained Transformer.

Other models work similarly to GPT-3: Eleuther AI publishes GPT-Neo and GPT-J. From Meta (the parent company of Facebook) comes OPT. From Germany comes Aleph Alpha. And new ones are added all the time.

The underlying principle for all Large Language Models is Deep Learning. The models read as much as possible during their training, ideally the entire Internet and every book. In this way, they teach themselves the thematic facts from the knowledge available worldwide. They also learn grammar and vocabulary in English, German, and basically every other language.

When they finish learning, they know what people in a similar situation would likely say. Just as Amazon would pitch statistically appropriate products, Language Models act in the sense that: People who have said „Hello!“ have also asked „How are you?“ afterward.

GPT-3 and Co: Is there a cheaper option?

Yes, there are cheaper alternatives to Jasper if you subtract all the content, which you can watch for free on Youtube. And if you forgo the community that restricts Jasper to paying customers. Then you can use the actual underlying artificial intelligence in other ways.

If you’ve learned what you can do with a Large Language Model from Jasper’s community and content, then you can apply that to other tools with a little bit of your own thinking.

Other resellers of Large Language Models include Longshot and Rytr. These and several more offer a free pricing plan—albeit limited. Other providers, such as Gocopy and Copysmith, offer a trial period. We’ll tell you more about what options you currently find in the market in another post in the coming weeks.

Large Language Models without Resellers are another way to get full AI power at a lower price. With them, you don’t have to subscribe; you pay by usage, pay-as-you-go. For many months from launch, GPT-3 was only available to a select few. In November 2021, Open AI opened up access to everyone—and is giving away $18 as a starting credit (that’s enough for many, many words). There are also many alternatives to GPT-3 that offer direct access. More on that soon, too.

Languages: More than just English

GPT-3 was trained primarily in English, and it shows. Jasper integrates DeepL, the well-known good artificial intelligence for translations, for better texts in other languages. This article as well was first published in German, my native language, then translated into English with DeepL (and some human editing). Since you are reading this version, I assume your mother tongue is not German. So, maybe you’d like to have content written in your native language, too? For me, this was definitely important.

That’s why I also tried Neuroflash. Neuroflash calls itself: Your #1 AI text generator for German texts. To do well in this language, Neuroflash puts a filter of current German-language texts on top of the output it generates with GPT-3. This process worked beautifully. It may be worth a try for you, though, as a similar high-priced alternative to Jasper, with examples and tutorials in German.

For me, a different tool is at the front of my testing roadmap because of the languages: I want to put Aleph Alpha from Heidelberg to the test next. It relies on its own Large Language Model, an alternative to GPT-3. It has been trained from the start in several European languages. For example, in German, my first language, I can judge the quality myself. Also in French and Italian. These are languages that I know to a greater or lesser extent. Our texts are regularly translated into these languages because some of my customers publish in them.

02 Assistance systems: Write better, faster and safer on your own

„You can’t edit a blank page,“ says US bestselling author Jodi Picoult, and she’s right. What’s important to me in the writing process is that I get a lot of stuff on the page quickly in a first run. Then, in the next one, I cross out a lot of stuff again and eliminate the mistakes. In this workflow, a few more tools help before and after. They keep you on track and make sure you reach your destination safely—the assistance systems.

Speech recognition and transcription

Speaking is faster than typing. So, why not use dictation for a change? AI has brought tremendous advances in dictation. Speech recognition is much better in 2022 than it was just a few years ago. I use Google’s solution on my Android and in Google Docs. Also, Apple’s onboard tools in the Mac OS X operating system. I will compare the tools more thoroughly soon. The most crucial difference: The Mac stops when it doesn’t understand a word—forcing me to input correctly. Google’s solutions just write anything, even if it’s wrong. That’s why it’s important to start editing soon after dictating.

Errors are part of speech recognition—not only with Google dictation solutions. Automatic transcription of recordings also generates many of them. Youtube automatically transcribes every video in the background, with recognition rates that sound high right off the bat. But even if it gets to 95 percent accuracy, depending on the language: It’s usually the other 5 percent from which the gross errors arise.

Transcriptions are, therefore, not useful as finished texts. But as raw material for writing, I wouldn’t want to miss them anymore. The combination of sound and transcription is unbeatable. With a transcript, the video or audio becomes browsable. That’s why I let artificial intelligence transcribe the interviews I conduct. Afterward, I can quickly jump from keyword to keyword and find great quotes.

This is working reasonably useful for German and English in the free Youtube-Studio version. Trint is also a good and comfortable option (alas expensive). If just English is enough for you: Loom does its job well, Otter is the market leader. But if you also want to transcribe Swiss German: Good luck. We have found only useless junk so far. I’m excited to test out Whisper from Open AI soon. It recognizes Scottish better than I do, anyway. If you find something, let us know; our gratitude is guaranteed.

Spelling and style check

Sometimes, I change direction in the middle of a sentence, and finish it with a verb that doesn’t match the preposition I started with. I also make a ton of typos with my unorthodox six-finger system. A spelling and style checker will help with both later. For me, what’s on board with Microsoft Word or Google Docs isn’t enough. I usually use Language Tool—but only after I’m done writing. Language Tool handles all the languages our team produces in (English, French, German and Italian).

For English as the original language, I’m currently testing Grammarly. The free version can do what I know from Language Tool. In the premium version, Grammarly’s features go far beyond that. Especially the style suggestions are super helpful for me. I speak English fluently. But unlike in my native language, I often don’t intuitively know if a sentence could be put more adequately.

Plagiarism check

Grammarly also has a plagiarism check on board, included in the premium subscription. Jasper offers pay-as-you-go services from Copyscape directly in the app.

For my own production, I don’t really need a plagiarism check. Most of the time, I can remember if I wrote the text myself. At least, that’s how it used to be. Today, so much text is created on my screen that hasn’t come from my own fingers. And this is where confusion can arise. Did this paragraph come from the pen of my artificial intelligence, meaning the copyright is mine? Or does it come directly unchanged from the research—and would still have to be rewritten?

Suddenly, the question of plagiarism also arises for solo authors. In the past, only publishers who worked with a whole team, including external freelancers, whom you don’t want to trust blindly, had to deal with this. If you publish such output, you have to check for plagiarism.

One would expect the texts penned by GPT-3 and Co. to be free of blatant 1:1 copying. But in the end, Large Language Models are stochastic parrots—trained to ape language patterns that other people have written before. Same same, but different—that is the goal. In this way, you can be absolutely certain, that very similar texts are produced, which is precisely the operation principle. At the same time, there is no ruling out that perhaps even clear copies are created.

So far, no single text written by GPT-3 has failed a plagiarism check. Has anyone experienced something different? Any relevant information is welcome.

Whether it’s a blog post, white paper, brochure, or video script: Without good content, you won’t get far in writing. Either you already have it in your head. Or you have to do some research. Artificial intelligence can help here, too. Just like a map helped you in the past—or a navigation system does today.

Research: Before and after writing

Whether I interview experts or write something myself, I can only do so if I know the subject. I don’t have to be an expert in the subject matter, but I do have to be familiar with it—or have read up on it for the occasion.

When I’m writing, I notice that I’m still missing information from time to time. Then I’m tempted to do some quick research. But that doesn’t end very well—More often than not, I end up in a web of my own distraction. If I want to be efficient, I can’t do any more research while writing. I have to know everything beforehand. So, that, during writing, I can put on blinders and maintain focus.

If I have a Large Language Model write content, this applies even more: I have to know a lot about the topic beforehand. I have to keep the artificial intelligence in check to ensure it does not lose touch with reality. I can’t trust that it will automatically make sense. GPT-3 and Co spread false information without batting an eye.

„When humans and machines play well together, great opportunities emerge.“

Abraham Lincoln

This quote is false—100 percent fictitious. In this case, the culprit is not even artificial intelligence. But my own human brain. I just wanted to exaggerate for a moment what could happen.

Abraham Lincoln taking a selfie with his smartphone (Midjourney).

Artificial intelligence strings together what fits well into the context statistically. And when it looks for the author of a quote, one of the most famous authors is always a good choice: Abraham Lincoln.

Whether with or without AI, I don’t want to deal with research while writing.
Before and after, I need the research all the more urgently. Before writing so that I know my way around the subject. After writing so that I know what’s right—and what’s missing.

Content briefs: a blueprint for every page

I regularly use Frase and occasionally Surfer SEO. Whether blog post or landing page: Both tools help to create a content brief before writing. To do this, they go to the topic on the Internet and evaluate what competitors write. Or, to be more precise, with which statements the competitors are successful—and make it into the top 20 on Google. Recently, such a function has also been on board at Neuroflash.

For me, a content brief is a good map of what might be worth writing about. And when I’m working with subject-matter experts—for example, on a client assignment—I quickly have an agenda of what I want to talk about with an expert in an interview.

If you want to abuse this power: You can hardly click together plagiarism faster than with the content briefs from Frase, Surfer SEO, and Neuroflash.

Frase has a Language Model for writing on board, which works fine in English (but didn’t convince me in German). Neuroflash is at home with the Language Models and, therefore, expectedly good, even in German. Surfer SEO teams up with Jasper for rewriting and creates an almost seamless integration.

One might be tempted to have the freshly clicked-together plagiarized content rewritten paragraph by paragraph. More and more people are doing this, putting these texts on their affiliate sites or selling them as SEO texts. This is no worse than cheap texts, priced per kilo from Textbroker and Co. However, content that inspires and offers its users added value is not created in this way.

Longshot also pursues the complete integration of research and writing. Here, the SEO ranking of the finished text is missing at the end. Instead, the process of finding a topic, researching it, and writing it merges frighteningly quickly. In the process, many words and functioning texts are created rapidly. Nevertheless, I can’t shake the impression that after reading it, you feel like after a visit to McDonald’s: full, but not really nourished.

Content Planner: The topics for the whole site

I refused to do SEO for a long time. I found the tricks of black-hat SEO repulsive since the noughties. Technical SEO is done by others anyway, whom I like to book on a project-by-project basis. So, my credo was and still is: content that people want to read is the best search engine optimization. Why even think about SEO then?

Today I know: SEO is important to plan what content I should write in the first place. For publishers, keyword research is the best market research.

I love keyword research. I often do content audits and rely on Google Search Console for each. It tells me what works—and what doesn’t—on an existing website. I love GPT-3 and Co for how they speed up an initial brainstorming session. (À la: Make me a list of 20 car brands. Or: Name 50 big German cities. Or: 100 popular French male names). I like the Related Keywords at Ahrefs and the Keyword Ideas from Ubersuggest; they quickly and reliably expand the ideas from brainstorming.

I hate keyword research. Just as quickly as the ideas come, I feel overwhelmed. Yes, Ahrefs and Ubersuggest tell me what the search volume is. That sounds as if I could then decide what I want to write about first, what would be worth writing about. But the numbers from the classic SEO tools apply only to a single keyword.

But when I write an article on one of the keywords, I automatically cover many other keywords as well—because they simply belong to the topic in terms of content. Clusters are what SEOs call semantic groups of keywords. But how should I group these keywords?

Tables, numbers, charts—with Ahrefs and Co, it looks at first glance as if everything is completely data-driven. But when it comes to grouping the keywords, suddenly everything is gut feeling again? That’s why I’ve always felt a bit left alone by the classic keyword tools of the SEOs.

Until, suddenly, artificial intelligence gets more out of that data. Simply upload a list of keywords and let an AI do what artificial intelligence is known for: It proceeds to recognize patterns and makes suggestions for suitable clusters using machine learning.

Cluster AI, Keyword Cupid (both with not-so-cheap subscription models), and Keyword Insights (with a pay-as-you-go model) are working on the principle of pattern recognition. All three make sensible suggestions and quickly reduce a list of hundreds of keywords to twenty or thirty suggestions for individual blog posts or landing pages—including aggregated search volume.

Zenbrief and Surfer SEO use another way of grouping. With Zenbrief, you can import your own lists from other tools instead of creating a new one. However, it does not look at all search results, but lets a language model decide which topics belong together semantically. The Content Planner feature of Surfer SEO works similarly. You don’t bring a list here, but specify a broad, generic topic. Whether Zenbrief or Surfer SEO: The groupings are partly questionable. Many article ideas are generated—in any case, that is an excellent basis for a discussion with experts on the client’s site.

Keyword Optimizer: No detail forgotten

I repeat myself: If I write an article about a keyword, I automatically cover many other keywords. Maybe even most of them. This happens by itself when I explore different facets of my topic. What never occurs automatically, however, is that I cover all keywords that people google for on the subject.

Some keywords will always be forgotten. Whoever wrote the text—myself, someone else in the team, or an artificial intelligence—it’s worth running an SEO optimizer after the first draft. For this feature, I only use software that I mentioned above for other features—for me, these are currently Frase and Surfer SEO.

Fact-Checker: Desperately seeking the truth

The faster you produce text, the more critical it becomes to check the facts. Whether you use artificial intelligence to paraphrase a lot of foreign content or create it from scratch, you need to know if the statements in the texts are factual or pure assertion. It would be nice if you could also rely on artificial intelligence to evaluate content. Here, however, there is still a considerable gap.

Longshot brings a feature called Fact Check. With this, you can check single sentences. The function shows precise results if the statement is true—then Longshot’s Fact-Check quickly finds different sources that claim the same thing. However, if the claim is questionable, the result is not apparent—the sources scatter widely. I hardly ever use this fact-checker because afterward, I am often just as clueless as before. The only clear thing, in this case, is that more research is necessary around this statement.

I wish I could have the facts checked automatically. However, I haven’t found a suitable tool yet. Do you know one? If so, I’m excited about hearing from you.

4. Gigafactories: Generate thousands of texts from data

When I write for myself, I write to organize the world for myself. Writing is work on the thought. Writing can take as long as my brain needs to understand the world. All the tools I’ve introduced you to so far help me do that. I can see different possibilities faster. Find solutions away from the first best thought. And find my mistakes.

If the goal was a single text, the work is finished.

In other cases, it only really starts after that. When I work for clients, I often write prototypes. Samples of teasers and headlines. Samples of blog posts or product pages. And then, it’s on to scaling those samples.

If more blog posts should be created, the process starts all over again: new topic, new content brief, new writing process. The same tools I introduced to you in chapters one through three of this article come into play again. Rinse and repeat.

Some texts, however, can be standardized even more. It’s worth it if you need a lot of them. Automation makes sense as soon as you want to produce 100 texts—or more: you can even produce thousands.

Typical use cases are product texts for products from the same category in the store, such as 100 different winter tires. Or 25 different TV sets in four languages. Or a current weather report every hour—that’s as many as 168 per week.

Scaling is required for such quantities. In the past, this inevitably meant a team of copywriters. Today, it’s time to look for a different set of tools. Data-to-text solutions such as those from AX Semantics or Retresco are in demand. These help to create text from data. This works great for anything that could be expressed in a spreadsheet.

For winter tires: How deep is the tread on this tire? What is the top speed it is rated for? Can you drive it in the summer? And for the weather report: How cold will it get at night? What is the maximum temperature? And how strong will the wind blow?

The artificial intelligence can then write a text from all this information. To do this, it packages the values from the table into nice sentences. And it can also classify the values: Is the temperature particularly high for a day in May? Then the algorithm will express that. And if the wind is neither particularly strong nor exceptionally weak, then a sentence about the wind strength would not be relevant and may be omitted.

In contrast to the Large Language Models, no sentence is created in the Data-to-Text solutions without a human developing a rule first. That requires effort before the first text can be created. If you are a copywriter, you basically program the algorithm. Then the machine can scale your ideas. It automatically generates texts—always correct, always unique, and consistently well-written.

And if you need one or more other languages, you don’t translate every single text afterward—you just translate the set of rules before the creation. This way, the texts are not only created in German or English but also in French and Italian (very important here in Switzerland) and many other languages. AX Semantics offers a total of more than 110 languages from all over the world.

Conclusion: Find the tool that fits your needs

I like to play with software. Writing is my profession. I combine one with the other, so it’s only logical that I try out many tools—all the way down the Rabbit Hole.

You don’t have to take it quite as far as I do.

I hope you found some tips in this article and now know which tools you want to try yourself. Take the first tool that appeals to you. Try it, then another, one tool at a time. In the end, keep two or three that you really use.

I regularly use premium resellers like Frase and Jasper. So in the medium term, I’d like to get rid of the subscriptions and pay-as-you-go only for the volume I actually use. For that, I’m going closer to the source, using GPT-3 directly, and just starting my tests with Aleph Alpha.

Without a reseller closer to the source: You can do that, too. Requirement: You know what you want to do. You can write the prompts for your use cases yourself. Or you know where to get the ideas for use cases—and the recipes for them.

What would help you? Which text types do you need most often? Which ones give you the most trouble? Let me know. Then I can write a detailed guide that will help you—and maybe it will appear here soon.

I appreciate the feedback. Talk to me if you have experience with any of the tools I’m mentioning here.

And if you want to talk to me about how you can make artificial intelligence work for you in copywriting—I’d be happy to talk to you on a short call, do a workshop with your team, or speak to your guests.

Cover image: A robot typing on a laptop, art station (Stable Diffusion).

About the author: Arne Völker trained as a journalist and was a screenwriter and copywriter. Today, he works for his clients as a consultant and helps them scale content—recently also with artificial intelligence. That’s how he becomes: “The copywriter who replaces himself.”

About the links: We sometimes recommend books, products, and services. These are always only those we use or will try out soon. Currently (as of September 2022), we do not receive commissions for this from any provider. Later on, we might include affiliate links—but even then, this attitude will not change.

About the translation: This article originally appeared in German as: Texten mit künstlicher Intelligenz: 25 Tools, mit denen du besser, schneller und mehr Content produzierst. This version is a machine translation done by artificial intelligence (DeepL, to be exact), curated with some affection by the humans of my team. If it gets a lot of traffic, we will revise it sentence by sentence with lots of love.