What Are HyperWrite Personas (& How To Use Them)

HyperWrite is already one of the more “human-sounding” AI programs on the market. Given that it’s an AI writing assistant, its default tone, voice, and style needs to be a level (or two) above generic generative AI. And the beauty of HyperWrite is that you can use its many, many tools to tweak and refine its output until it sounds exactly as you want it to be.

But what if you could skip the post-process editing and get it to sound a certain way, straight out of the gate?

HyperWrite’s newest feature called Personas essentially does just that! It allows users to program the AI so that it produces specific, tailored results on the first try. Given how customizable they are, Personas are shaping up to be yet another powerful and intuitive addition to HyperWrite’s suite of professional writing tools. 

‍

Here’s why.

What Are HyperWrite Personas

The dictionary definition of ‘persona’ reads; “a role or character adopted by an author, actor, etc. or in a game.” And, in a way, this definition works for HyperWrite Personas, too.

HyperWrite Personas are basically a writing style, voice, preference, and personality that your AI assistant can adopt. In doing so, HyperWrite’s output will change to fulfill the specific parameters set by the persona. It’s like giving the AI clear, strict instructions on what to do and how to write before every interaction. 

Different Personas, Different Results

If you’ve used AI to write content for a while now, you might have noticed a few of its favorite words: ‘delve’, ‘tapestry’, ‘testament’, ‘embark’, ‘journey’, ‘fast-paced’, and so on.

With Personas, you can set it so that HyperWrite never uses these words again.

If you use HyperWrite for academic or technical content, you may have gotten used to asking it to do a dozen things before it writes. Like fact-checking statements using Scholar AI, for instance. Or always including reported statistics and/or quotes from peer-reviewed journals when applicable. 

With Personas, you no longer have to ask. You can set it so that HyperWrite does that in every interaction, regardless of what you ask it to write. 

You can also seamlessly switch between Personas. Your writing can sound like an enthusiastic, well-traveled blogger for one iteration and a serious, slightly dry academic writer for the next. You can even create a Persona that will only respond in Spanish!

Under the Hood

HyperWrite is one of the first companies to serve a custom Llama 3 70B, and it's using it to power these Personas – with incredibly satisfying results. This is one of the most consistent, natural, and human-like models to date. And we can only expect its performance to improve over time.

Truly, HyperWrite Personas might be the cure for generic AI writing. 

How Do You Use Them

HyperWrite has always excelled in accessibility and user-friendliness. Their tools are easy to use and interact with. You don’t need to have any sort of background knowledge or technical experience in coding, programming, or AI integration. The tools always have instructions and some sort of example to follow.

HyperWrite’s other major services – AI agents and, now, Personas – are the same way. They’re easy to access and fairly easy to figure out.

Let me walk you through creating your own HyperWrite Persona.

Step 1. Browse Personas

Go to the HyperWrite Personas page. This page lists all the current Personas you can use. 

It’s also where you can create and edit your own custom Personas.

Step 2. Create New Persona

There should be a blue button on the upper right-hand corner of the page that says + New Persona. Click on that to create your first HyperWrite Persona! 

‍

Step 3. Define Your Persona

After clicking the “+ New Persona” button, you’ll be taken to a page with the following sections: Instructions, Examples, and Background.

‍

‍

These three categories are what you’ll use to define your Persona. Let’s go through each one.

I. Name & Instructions

First of all, name your Persona. This will help you identify it immediately once you have a fair stack of Personas to choose from. A name will also help you determine what the Persona is for.

‍

For this example, we’re going to make a Persona that writes dedicated professional software product reviews. So I’m going to name it “Sally the Software Product Reviewer.” 

‍

Let’s say I have two other existing Personas: Travis the Travel Blogger and Anne the Academic Paper Writer. By giving these Personas distinct names, I’m able to differentiate them and determine their function immediately – as opposed to vague or generic labels, like “Persona01” or “05.22.2024_Model.”

‍

‍

After the name, you’ll need to give your Persona custom Instructions. 

‍

Be as specific and as detailed as possible for this section. These instructions will set the parameters for how your Persona writes, acts, researches, etc. In other words, this is where you’ll inform your Persona of their role, their voice, their focus, their thought process, and so on.

‍

Some things to consider when giving instructions:

‍

• What is the Persona’s role? What do they do?some text
  • Example: software product reviewer, travel blogger, teacher/tutor
• What kind of vocabulary does the Persona use?some text
  • Example: conversational/every day, technical, academic, highfalutin
• What is the Persona’s voice?some text
  • Example: friendly, professional, authoritative, hyped/salesy, strict

‍

Don’t worry about potentially missing or miswording instructions. You can always modify them later on. But do try and list as many parameters and descriptions as you can think of to optimize your Persona’s first draft.

‍

II. Writing Examples

As with the custom Instructions, you’ll need to provide as much data as possible for the Writing Examples. If the Instructions set the parameters for how the Persona acts, the Writing Examples are the Persona’s references. In acting, they’re the scripts; in public speaking, they’re the cue cards; in academics, they’re the textbooks.

‍

These examples are what the Persona will copy – and eventually emulate – every time it writes something. It’s one thing to tell the AI to “write in a friendly, conversational tone.” But it’s another thing entirely to have a reference (or two) of what that “friendly, conversational tone” reads like.

‍

‍

Think of haikus. If you’ve never written or read a haiku before, it’s difficult to imagine what a poem with a 5-7-5 syllable structure looks like. Even though the instructions are clear, conceptualization can be tricky. How do you divide the sentences, for instance. Should you limit one idea to one line? Should the last words rhyme?

‍

Therefore, it’s always better to have some sort of reference to follow. In the case of HyperWrite Personas, the more references you give it, the more accurate the AI will be in producing the writing style you want it to follow.

‍

‍

For this reason, I suggest providing three or more Writing Examples. You can provide as many examples as you want as long as it’s below the 10,000-character limit. 

‍

‍

Adding more writing examples is easy, too. Just click the bright blue + Add example button in the upper right, next to the Writing Examples heading and just below the character limit bar.

‍

‍

And when you’re done writing or pasting the text you want the Persona to copy, don’t forget to click the blue Save button. 

‍

‍

III. Background Information

This section is where you’ll provide the Persona with information that it can reference when asked specific questions. For instance, the URL of the website that the Persona will post to. Or the name of the company that the Persona works for. 

‍

‍

The idea is to give the Persona immediate access to relevant data that it should prioritize when asked. With the Sally example, I reiterated the Persona’s name and its purpose. 

‍

‍

I also instructed Sally to specifically inform a user that they cannot review anything that is not classified as a software product. This way, the Persona has a default (negative) response to give when asked to critique anything other than. 

‍

Without this limitation in place, there’s a chance that Sally will either (a) write reviews for any type of product – regardless of the classification – or (b) simply not respond when given a non-software product to review.

‍

Here are quick examples of Sally immediately retrieving the specified information when prompted. 

‍

When asked what Sally does:

‍

‍

When asked to review a non-software product:

‍

That’s A Wrap!

And in three simple steps, you’ve made your own HyperWrite Persona! You can take it a for a test run by clicking the blue Test Persona button on the left side of the page:

‍

HyperWrite Personas vs. Default: The Difference

Now that we’ve got Sally all setup, let’s see if there’s actually a difference! 

‍

If you head back to the main HyperWrite Dashboard, you’ll notice there’s a dropdown menu beneath the right side of the general AI chatbox. It should be set to HyperWrite by default. 

‍

‍

If you click on the dropdown menu, you’ll see a list of your available personas – including HyperWrite’s default options (Academic Researcher, Wild Card, etc). Select the persona you want to use. 

‍

In this case, we want to use Sally.

‍

If you can’t see the persona you want, select the “Manage Personas” option. It should take you to the Browse Personas page where you can see – and choose from – all your personas. 

‍

‍

Click the blue Use button to switch to that persona. It will automatically reload you back into your HyperWrite Dashboard. Check the dropdown menu to confirm that you’ve switched to your desired persona. 

‍

TESTING: PERSONA SALLY

Let’s test the difference with a short Google Docs review. Here’s Sally’s prompt:

‍

‍

And here’s her response:

‍

‍

Immediate Observations:

‍

• Formatting – the response is cut up into short and long paragraphs for improved visual rhythm and readability 
• Feedback – the review mentions drawbacks but immediately balances them out with positive points
• Flow – the flow feels natural; the jump from one paragraph to the next doesn’t feel too sudden and the points covered in each one are not repetitive

TESTING: PERSONA HYPERWRITE (DEFAULT) 

Now we give the same prompt to HyperWrite. 

And here’s the review:

Immediate Observations:

• Formatting – the review is one long paragraph; no cuts or divisions for easier reading; very different from Sally’s review
• Feedback – the review mentions drawbacks but immediately balances them out with positive points; similar to Sally’s review
• Verbiage – this review reads a little more poetic; it feels like there are more adjectives than necessary

THE VERDICT

I think it’s fair to say that both reviews don’t look or read the same. The formatting alone is a clear difference, sure, but even the tone and verbiage are dissimilar. Sally’s review reads more direct and user-focused. HyperWrite’s review highlights the features in a descriptive, almost salesy, way. 

Remember; this isn’t a question of which one is better. Both reviews, for all intents and purposes, are fair and balanced. What we’re trying to test here is whether creating and using a HyperWrite persona significantly changes the AI’s output or not.

And it seems that it does! 

HyperWrite Personas Are Changing the Landscape of Generative AI

Honestly, HyperWrite Personas are poised to be potential game-changers for generative AI. They’re accessible and easy to use, and you don’t need programming experience or expertise to make one. This opens the door for widespread adoption by anyone – not just AI experts!

They also offer an elegant solution to the often robotic and impersonal tone of general generative AI platforms. And as far as use value is concerned, HyperWrite Personas are both ethical and efficient. They leverage human examples and input, and they’re only as capable (and as limited) as they’re programmed to be. 

This means you get all the benefits of AI but with clear boundaries and ethical considerations in place. 

As AI continues to evolve, AI Personas are shaping up as the perfect bridge between cutting-edge technology and natural human communication – a win-win for both efficiency and responsible development.