Do you still recognise you?

Do you still recognise you?

It looks as though everything is starting to look like a big blur. A thick pumpkin soup where everything went through the blender. I used to be able to remember who wrote what on the internet. Now one message bleeds into the other. Has become interchangeable. No honest recognisable voice left. Everyone seems to be a wordsmith nowadays, but few are able to condense their message into few words. This is happening while attention spans have been failing for years.

One could argue that this should drive companies and its leaders to understand they need to make sure their message is uniquely positioned in their way of writing. True to their brand identity. And that less is more…

What is left is paths of writing that will draw canyons across the communication landscape. From the emdash, to the pointless use of the ¨rule of 3¨ and the ¨why it matters¨.

Maybe when writing becomes so easy that is an afterthought. It makes you you forget that writing is thinking and if you do not suffer your way through it, you will not be able to judge the outcome honestly. Stumble upon thé ideas that actually lay deeper beneath the surface, awaiting to be discovered. If not to abandon the idea in total as it turned out not to be a worthwhile pursuit.

Do you wonder what your (potential) customers are feeling? When you are not willing to go the extra mile to craft your own messages but resort to a statistical machine. To shortcuts. That surely allows you to mass up your presence in amount of posts. Yet lack any other form of presence that people so much grave. The one of honest intent, attention and Authenticity. The one that builds trust.

Lets have a close look at what is happening

Something has triggered me in the change i am experiencing in the online communication landscape. It has to be because when working on concepts for clients every word matters and is measured. Does it add to the story or does it break it if not even worse, cause distrust.

In my infinite thirst to keep learning every day i have been digging into these llm models to understand what is their foundation. This is when my interest was sparked on how these models are impacting the communication landscape i started looking for patterns and digging deep into research articles. These articles articulated precisely the things i was already sensing.

I will be quoting from 2 of these articles below that give a neat overview of what is happening.

1. MP´s are almost certainly using ChatGPT to generate Commons speeches

Apple Tokamak dug into the word of MPs in the UK and their Commons speeches word use over time. They went a bit further than just words, they looked into patterns often expressed by LLMs (AI). They found compelling evidence that directs to the idea that these speeches are more often now created or generated with the help of LLM´s (AI).

The graph below showcases that certain use of words and patterns, of which we are aware that they are LLM´s favourites, has exploded.

The second graph shows modern word-use of social platform seeping into the landscape of PM speeches. Where this trend already started before LLM use it definitely showcased explosive growth after the introduction of LLMs.

In the article there is a deep-dive into a few speeches. I must say, there is highlights there that once seen are hard to un-see. For instance the places where LLMs want to add defined patterns to sentences and enforce that structure at all cost (rule of 3). These speeches, at first sight, look properly written. But when put under magnifying glass they fall apart.

In my experience i have seen these patterns appear as well while working with llms. And it is hard work to keep them away from overcomplicating sentences, often bloating the paragraph with non-sensible formulations. Even though at first glance looking like professionally orchestrated parts of content. These generations are often above my level of writing and maybe herein lies the problem. Because the word smithery is of such level it becomes more and more difficult to catch it in the act of delivering empty slob. 

When you have not put the actual effort into crafting the raw material and work your way through discoveries… how will you be able to judge the material presented to you? Even more so, multiple iterations of it… 

2. ChatGPT Is Changing the Words We Use in Conversation

Hiromu Yakura, a postdoctoral researcher at the Max Planck Institute for Human Development in Berlin started noticing changes in his own spoken word use since using ChatGPT for all kinds of jobs. This sparked his interest into researching it on a broader scale. His study shows that in the statistical patronised way these llms (AI) works seem to be creating a feedback loop back into our spoken language.

“The patterns that are stored in AI technology seem to be transmitting back to the human mind,” says study co-author Levin Brinkmann, also at the Max Planck Institute for Human Development. In other words, a sort of cultural feedback loop is forming between humans and AI: we train AI on written text, it parrots a statistically remixed version of that text back to us, and we pick up on its patterns and unconsciously start to mimic them.

He is highlighting that there is risks that we are currently not taking into consideration. One of those risks is the singular nature of the AIs our diversification of language is under threat. We humans do of-course copy each other to a certain extend, but we are still diverse in our being. One could say keeping some equilibrium and more organic growth of language.

Another risk i would like to add here is the moment it will start to train on the new materials being created and conversations minutes. It wil be as if a snake is eating its own tail. What will happen then? Will it start to draw deeper grooves into these patterns? And what does that mean for our ways of communication? Our Culture? Our understanding? I am sure we will have a look at that in a next article.

Given that ChatGPT has changed how people talk just two and a half years into its adoption, the question becomes not whether AI is going to reshape our culture, but how profoundly it will do so.

Do you still recognise you?

Taking the above as a clear sign of change in communication style, word and pattern use. The question is while looking at your own writing being augmented by LLMs(AI), is it still you? When you start using words outside of your usual vocabulary. Shaping patterns for the sake of patterns. And if you can no longer identify you, who will?

Wouldn’t you agree that the point of writing and posting on for instance Linkedin was to become recognisable?

What does this mean for the future

Seeing this pattern begs the question what that will mean for the future? What does this mean for the relationship with each other? How do you feel if the content you receive feels as if there was no value in communicating with you and thus no effort made.

Furthermore there is evidence that people who generate their content have limited memory of what they have created. Which is maybe not so much of an issue when you are providing support tasks. But when you are an actual PM of a country that becomes a questionable issue. It begs to question: ¨what is their intent?¨ and why should we trust?

Brand voice matters

Brand voice mattered always. And your actions define your reputation. The current market gives the opportunity to be different, to stand out and be BOLD. We are in a era where trust is crumbling. This is where you could stand out. Be personal in your communication.

All this does not mean ignoring AI. But knowing where it can be helpful and augmentative. And where it is diluting. We at We Are Bold use AI every day but remain highly critical and selective where. 

For this article as an example i used an LLM to criticise it. To see where there is lose ends. Which I would be able to do if I take a good amount of distance from the writing and coming back later. Now many of these easy fixes can be done already while looping through the article the first round. But i feel like the wording is up to me and should reflect me.