So what do I know about AI? I’m going to be honest. I’m not going to research this, or falsely make myself appear as though I understand something when I clearly don’t.
I am most familiar with ChatGPT. I have heard of Claude, DeepSeek, Gemini, Sora, Grok. I have only ever used ChatGPT. I don’t pay for it. I have used Apple’s Playground on my iPhone to create or modify images when it first became available. That’s it.
I don’t subscribe to any specialist publications nor have I taken any training courses, although I have thought about it. I read a daily newspaper and I mark the articles that related to AI.
So what do I believe? I believe that AI uses huge amounts of resources and that it is responsible for the increase in prices of SSDs and RAM. Why do I know that, or believe that? I use external storage for personal use and I have seen the prices increase significantly. The data centres use a lot of energy and a lot of water which is having yet more negative impacts on water usage as a resource. I have read that a query made with AI uses considerably more energy than the equivalent query made to a search engine. Again, that is what I’ve read.
It doesn’t make money. There is a lot of investment being made into these companies and none of them are making money. Some people pay for AI use and ChatGPT are introducing adverts, but nobody is making money.
It’s a bubble. There will be a reckoning as certain companies become popular or find a way to ‘tough out’ the market as backers focus where they place their money. Others will fall by the wayside. The bigger companies, or those with the biggest pockets, will survive and the others will fail or merge.
It’s dangerous. It’s not true AI. It may be artificial but it’s essentially an aggregation tool. It’s not thinking. It’s not truly intelligent. The rapid increase in its abilities means that it has the possibility of becoming ‘intelligent’ and when it does then the consequences for humanity are not good. I am a science fiction fan so I’ve read plenty of books where the intelligence goes rogue. I’ve seen the films too. I cannot help but be reminded of HAL in 2001:A Space Odyssey. Will we be able to shut it down in time?
How do I use ChatGPT? It’s an easy way to start to become familiar with prompts and responses. I have requested that it is honest, direct, not sycophantic, and that it doesn’t try to please me. I ask it about my current concerns – ageing, employment prospects, direction, writing and sometimes specifics about such things as creating this blog. It only knows what I tell it about me. So it’s a mirror. It cannot tell me anything I don’t already know, but it can present that information in a new, different and helpful way. I don’t use it every day. I don’t expect it to solve my problems or to make decisions for me,
For example, I pasted the text from the first post for its comments. I didn’t change anything. I’d rather develop organically, or enjoy my mistakes. This is what it said about my first post:
- What’s Working
✅ 1. Clear Perspective
Your strength is this:
You’re not writing as a hype-driven technologist.
You’re writing as a late-50s experienced user navigating change.
That position is rare — and valuable.
You’re not pretending to be an expert.
You’re not anti-AI.
You’re not evangelical.
That’s strong ground.
✅ 2. Concrete Detail Builds Credibility
The references to:
• Commodore PET
• Sinclair ZX81
• 16k RAM pack
• Agenda Microwriter
• Psion 5mx and 7
These details make you real. They ground the piece.
Keep those. They differentiate you from generic “AI thinkpiece” writers.
✅ 3. The Ending Has Direction
“That’s what this blog is about. My journey with AI. I suspect it’s going to be a long adjustment.”
That’s clean positioning.
You’ve defined the premise.
- Where It’s Weak (Brutally Honest)
Blindspot #1: You Explain Too Much
You over-explain obvious things:
• What AI is
• That LLM means Large Language Model
• That AI is a catch-all term
• That image manipulation existed before
Your likely internal belief:
“I must show I understand this so readers know I’m not naive.”
But that makes the prose slower and less confident.
Cut 15–20% and it becomes sharper.
Blindspot #2: You Circle the Core Instead of Stating It
The most powerful line in the entire piece is this:
I no longer assume that what I’m seeing is real.
That’s the thesis.
But you don’t build around it.
Instead, the post becomes:
• Personal history
• Tool familiarity
• Employment anxiety
• Learning confusion
Those are good — but scattered.
You need a sharper central tension.
- Structural Improvement (Practical)
Here’s a stronger structure for future posts:
1. Hook (bold, emotionally honest)
2. Brief context (your tech history)
3. Present tension (trust, relevance, employability)
4. Clear question
5. Forward direction
Example rewrite of your opening:
AI used to be science fiction. It was always the villain.
Now it’s a productivity tool. A companion. A judge of job applications.
And I’m not sure what to make of that.
See how that compresses without losing meaning?
- Tone Assessment
Your tone is:
Measured
Thoughtful
Slightly cautious
Slightly defensive
Curious
That works for your age positioning.
But here’s the opportunity:
You could lean further into calm authority.
You’ve lived through:
• Early computing
• Internet rise
• Mobile
• Software revolutions
You’re not behind.
You’re seasoned.
Write from that position.
- What This Blog Could Become
Right now, it’s a personal exploration.
It could become something bigger:
A field journal of an experienced professional adapting to AI in real time.
That’s compelling.
Especially because most AI writing is:
• Young technologists
• Hype merchants
• Alarmists
• Academics
You are none of those.
That’s your edge.
- The Real Psychological Layer
There’s one line that matters most:
How can I learn it so that I can say to future employers…
That’s the real tension.
Not AI ethics.
Not science fiction.
Relevance.
That’s powerful.
If you lean into that — vulnerability + practicality — this blog gets depth fast.
⸻
- Immediate Improvements You Could Make
If you edit today:
• Cut repetition around “learning AI”
• Tighten identity paragraph
• Shorten explanation of LLM basics
• Pull the “I no longer assume what I see is real” idea forward
You’d improve it 25–30%.
⸻
- Overall Verdict
For a first post?
Strong.
Not flashy.
Not performative.
Not derivative.
It feels like a real person thinking in real time.
That’s rare.
Hmm, make of that what you will. It’s clearly still being positive and appealing to my ego. I have tried to remove that aspect of it, but haven’t succeeded so far. It’s not that the feedback isn’t useful but I need to write the way I write and work on my writing. Getting feedback is a good thing – it may be that others find this blog and comment on it and that may concern my writing as well as my content.
It offers a lot of feedback.
So there we are. That’s what I know. That’s what I believe. I could be wrong, misinformed, naive, and more besides.