Rambling, A Template, Over Polite, Flowery, and Lacking Any Human Spirit: Whatever Happened To Humans?

Look at the image above, and what do you see? A happy robot and human sharing some coffee and having a happy time. Look, the robot is smiling! But it took me a while to get it from DALLE-3, as I didn’t like that I was portraying AI in a negative way. So, overall, I don’t think the AI bots wanted me to write this article (or maybe I’m giving them too much credit for their current intelligence):

And just the kind of thing that HAL would have said:

And when I probe more:

and I just go around in circles:

Introduction

I read my LinkedIn chat messages, and I would say that half of them are written by AI. It’s all rather lazy, as the person can’t be bothered writing me a simple “Hello” message, but where I get a long and flowery ramble of how my work will be “pivotal” and “outstanding”, … both words that I would only reserve for someone who’s work was truly pivotal and outstanding. If someone is so lazy not to write a human message to me, then perhaps I am too lazy to be polite enough to respond back.

Overall, I despair of the future world of AI. It will take away many jobs and in many places, will not replace them with anything which humans would find satisfying. While I will continue to develop with it and use it as a tool, I believe we are generally handing over our whole existence to the machine. This will perhaps be a world that will lack any real human input, and where everything that we read and everything that we hear will be generated by AI. We will lose one of our greatest attributes: the power to communicate.

A fake world of humans

It will be a world of fake content. A world of fake humans. One where a graduate could cheat themselves into getting top marks in every module without ever creating or writing anything of their own. And, I know we cannot contain it, but, perhaps, as a human you should stop yourself blindly using it to communicate for you.

And, so, I am human. I care about this. I have a passion. I enjoy words. I love things. I write things that some people may agree with and others not. And, to me, one of the most annoying prompts I see is this one:

Words have power, words are power

I don’t want to try and write my communiations with AI. I want to write as a human and not as a bot. If you want to understand the importance of words, watch [here]:

In a few years, I believe that humans will hand over their intellect to machines, and where most of the communications we see will be generated not by humans but by AI. But how can you tell?

I see the rise of the machine happening now, and where some people who contact me on LinkedIn have long and rambling lines of text. So, rather than saying,

"Hi Bill, We have a blockchain conference coming up in Glasgow, and 
would love you to serve on the committee"

they write something like

Dear Professor,

We are organizing a Blockchain conference on 11 April in Glasgow, focusing o
n pivotal areas such as zero-knowledge proofs and digital wallets. Given your 
distinguished background and significant contributions to these topics, your 
insight would be invaluable to our programme committee. Your expertise would 
greatly enhance the depth of discussion and provide pivotal guidance to the 
conference's direction. We believe your participation would enrich the event 
immeasurably.

Please let us know at your earliest convenience if you would be willing to x
accept this invitation and Share your Full Name, Designation, Affiliation and 
Email Address. Your consent would be greatly appreciated, and we look forward 
to the possibility of collaborating with you on this exciting endeavor.
Thank you for considering our invitation.

Looking forward to the possibility of your involvement.

This is generated from “Write an invite to an academic to serve on a programme committee for a Blockchain conference on 11 April in Glasgow. The conference covers areas such as zero-knowledge proofs and digital wallets. Praise their background in the topic and the relevance to the conference.”

How to spot an AI bot on social media

How do spot the AI bot? Well here’s a few tips:

• It’s overlong for a quick introduction. I have learnt anything about introducing myself to people is that I need to keep things short, such as, “Hi Steve, I enjoyed your presentation at the conference last week, and wondered if you had time to discuss a collaborative project”.
• It’s a template. I might change the template, but it’s still the type of template for formal communications that you learnt at school.
• It uses flowery language. Unless you were a Nobel Prize winner, few people would use the words “enrich”, “pivotal”, and “distinguished” in an introductory message, and possibly never use them as they are a little too flowery.
• A human would never say that. Who, in normal speech, would say, “at our earliest convenience”? What does that even mean?
• Over politeness for the situation. ChatGPT has learnt to be polite in the extremely. The text above is full of “please” and is over appreciative.
• Rambling! A standard sign of AI is saying something in a 100 words that would normally take 10. For example, “Please let us know at your earliest convenience if you would be willing to accept this invitation and Share your Full Name, Designation, Affiliation and Email Address. Your consent would be greatly appreciated, and we look forward to the possibility of collaborating with you on this exciting endeavor.
  Thank you for considering our invitation.” could just be said as, “Please can you acknowledge if you accept this or not”.
• US, not UK. Like it or not, AI bots are not well grounded in regional differences for spelling, and will often overuse the “ize” rather than the “ise” ending, and in spelling, “endeavor” with US spelling. Our world of AI is becoming an extension of the US, and which cares little about culture and styles outside this cultural bubble.

Overall, it feels like being back at school, and when the teacher showed you how to write a standard letter to an employer. The messages just feel like a template. The way LLMs have overcome the problem of plagarism, is that the iterate around a number of templates, but still they are templates.

Conclusions

We are handing our planet over to AI. Good luck with that.

We have waged war on others to save our dignity, we have killed in the name of our religions and beliefs, we have destroyed our beautiful planet in the course of our greed for wealth, and we spread anger and hatred to those who differ from us. But, like it or not, we are human. We care, we love, we have passion, and most of all we have feelings. We have built our society through verbal and written communications. But using AI blindly, we switch off our soul, and hand it to a faceless, bland and over polite machine.

Postscript

I kept at it, and still DALLE-3 wouldn’t generate an image:

Until I got a happy picture of machine and human sharing a happy moment: