How to Elevate Your Digital Discourse
A Guide to Reducing Emoji Overuse and Improving AI-Generated Content
In the age of social media, it's become increasingly easy to communicate with others and share our thoughts and ideas.
With the rise of AI language models, like those used by ChatGPT and Google Bard, we have even more tools at our disposal to help us craft compelling content that resonates with our audience.
However, as with any tool, there's a risk of misuse. In this post, I'll explore the bizarre phenomenon of emoji overuse and how to improve the quality of AI-generated content by editing it properly.
Yes, some people can tell when you’ve used an LLM (Large Language Model). No, it doesn’t matter. If your writing style suddenly goes from Berkeley to BuzzFeed (or vice versa), it’s noticeable. Don’t worry about it though, learn from your mistakes and elevate your digital discourse to that of a human being.
The Pitfalls of Emoji Overuse
While emojis can be a fun and creative way to add some personality to our communication, they can also be overused to the point of absurdity. A post filled with too many emojis can be difficult to read and can detract from the meaning.
Here are some tips to reduce emoji overuse:
Use emojis sparingly:
Instead of peppering your messages with a plethora of emojis, choose one or two that convey the sentiment you want to express.
Let your words do the talking:
Don't rely solely on emojis to convey your message. Use descriptive language to paint a picture for your reader.
Try emojis as bullet points instead:
If you want to use a specific emoji. or related emojis, they can work well as bullet points for longer form content, in paragraphs. I would avoid them for shorter form content though. If it's a list, try using numbers instead.
As a firm believer in the power of language, I find myself at odds with the ubiquitous emoji culture that has taken hold of our digital communications.
Don't get me wrong, I’m not entirely opposed to the use of emojis, they can add a touch of humour, emotion or irony to a message, and let's be honest, who doesn't love a well-placed heart emoji from time to time?
I purposefully avoid reading posts full of emojis. You haven’t grabbed my attention, you’ve lost it. It’s not that I’m a joyless curmudgeon who can't appreciate the finer points of visual expression (although, that may also be true). No, my aversion to excessive emoji use stems from a deeper philosophical conviction: namely, that emojis represent a kind of linguistic nihilism.
Think about it. Emojis are essentially a form of shorthand, a way to communicate complex emotions or ideas through a pre-packaged set of symbols.
Relying too heavily on emojis, we risk reducing our language to a series of images. We don’t want to go backwards, grunting and pointing at things instead of using our words to express ourselves. In the same way, relying too heavily on AI-generated content might also come with risks relating to language.
Improving AI-Generated Content
Tools like ChatGPT are incredibly powerful tools for generating content, but they're not perfect. The kind of prompt you've written and the keywords you've used are often clear to those in the know, especially when you share a piece of content full of random emojis without editing it.
In order to make the most of LLM capabilities, it's important to edit their output properly. If you don’t know how, ask (or hire) someone who does.
Here are some tips for editing AI-generated content:
This might sound a bit harsh but if your output really isn’t up to scratch, there’s something wrong with your input. Fix your prompt and try again. If you’ve hit with a wall of emoji ridden text, write a new prompt.
Read it out loud:
When editing AI-generated content, read it out loud to ensure that it flows smoothly and makes sense. There tends to be a lot of unnecessary repetition which you’ll pick up on if you read it out loud.
Use a human touch:
While AI-generated content can be impressive, it lacks the human touch that makes writing truly compelling. Add your own personality and voice to the content to make it more engaging, think human-centered technology.
AI-generated content can contain errors or awkward phrasing, so it's important to proofread carefully to catch any mistakes. Even little errors are easy to spot, for example a person who would ordinarily write in British English has now decided to write as though they’re an American.
By following these tips, you can elevate your digital discourse and create more compelling content that resonates with your audience. While AI language models are incredibly powerful tools, it's important to use them responsibly.
Make sure the human touch in your writing is not neglected by over-relying on LLMs and while I'm not opposed to the occasional use of ‘engaging’ emojis, I think we should all be careful not to let them take over.
Let's remember the power of words, and use them to express ourselves in all their rich, nuanced glory. With a little effort and attention to detail, you can make the most of these tools and take your digital communication to the next level.
Human-centered Technology is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.