I wrote yesterday about how to plan and extend my campaigning work. That post attracted quite a lot of interest.
What occurred to me after producing that post was that one of the easiest ways for my message to be amplified is to ask others to share what I am saying by quite literally amplifying my message in social media, but with tailor-made reposts, comments and even derivative articles or letters rather than simple retweets, likes or whatever.
It's my opinion that AI can really help this process and most people now have access to those, whether via Google Gemini, Microsoft CoPilot, or ChatGPT versions 4.0 or 5.0.
This might sound intimidating. It isn't. AI is a tool, and is one that can help anyone who wants to campaign, write, or communicate more clearly. Used with care, it can save time, and help you share your ideas.
So, my question is, would it help if I produced a short series on how to use AI for this purpose? The aim would be to:
1. Explain how AI works .
2. Explain how to write an AI prompt.
3. Work through the development of a prompt campaignmers might use to, for example:
- Write an X or Bluesky post or thread;
- Write a Facebook, LinkedIn or Subtack post, or
- Write a letter on an issue.
4. Explain how to refine the prompts overtime, using AI itself to assist this process .
Let me know.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
There are links to this blog's glossary in the above post that explain technical terms used in it. Follow them for more explanations.
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
Buy me a coffee!
Yes I think that would be a great help to explain how to magnify your message on social media by using AI
Thanks.
Yes please! And engaging in such exercises might also have the benefit of consolidating my understanding of those areas of economics you often touch upon and which sometimes leave me a little confused- bonds and gilts come to mind- areas with which you have a whole careers worth of experience whilst I much less so!
Noted.
Yes indeed….let’s build this thing!
Yes please!
Absolutely yes
Yes!
Yes. Please do. A practical guide to getting usable material from AI would be of broad general value to your readers, not just for campaigning. Your blog soon after Chat GPT’s launch, showing its authoritatively phrased history of a non-existent East Anglian railway, was a valuable warning about AI hallucination, and now I assume anything written in ChatGpT’s distinctive style is wrong, unless it is something I already know anyway. So, being able to write coherent English unassisted, I have avoided using it. However, the blogs you declare to be directed and edited AI do not seem to be of lower quality than the ones you do all by yourself, so you must have useful insights to teach us all.
It takes effort to make it work, I promise you.
I will be concentrating on producing short pieces.
I think it could be extremely helpful especially in HOW to deliver the message more effectively and personally.
The more you look at how mis and disinformation is used, they fuel survival bias with false positives. It gives people the easiest and safest way to make a decision, especially when counter to their benefit. So any new/science based fact can be muted with fear of failure this would take away everything they already have. Like walking in the woods at dusk with a friend and they yell “snake!” every time you see a stick. Better to always think snake than stick. I have been using AI to understand how to better position what really is the snake. For example when people give disinformation on coal here in the US, I have learned to leverage “Pick which plant we will open a mile from you; coal, nuclear, wind; Which affects you the least?”. I usually get silence or a version of build it in isolation.
🙂
Yes that would be helpful. I am a bit afraid of it!
🙂
“Write a letter on an issue.”
Overall I agree. Just a note of warning: the letters may have similar styles, A.I. is nothing if not unoriginal in the way it expresses things. In turn this can lead to e.g. politicos, ignoring the letter as yet another A.I. generated piece of dross (in his/her view).
Just saying.
I can never get it to repeat itself.
And I have tried.
Thats a big yes from me.
I am a software engineer that has been integrating AI into my workflow in recent months. It has been difficult. AI has significant limitations.
I regularly write to my MP. I just like annoying her! I use AI to try to construct the arguments and present it in a more “well written” style. It takes some significant “cut and paste”, editing and rewording to do this so if you have worked out something better, I would be very grateful if you can share it!
Let’s see.
That’s an excellent idea and I welcome your support.
I question, are there some people that simply don’t have the time or frankly, the patience, to unpack the economic jargon they hear, yet they’re the ones most affected by the policies that come from it. Perhaps we can help increase your reach to these people by asking simple, thought-provoking questions, encouraging them to think for themselves and realise that the economic “truths” they’ve been told aren’t necessarily facts.
Those people with less time may be attracted by shorter, punchier formats.
yes this would be a great help
I have so far avoided doing anything with AI and ChatGPT. I have a (possibly unjustified) confidence that I can write grammatical English and make my points succintly. I don’t want my style to be anonymised and my thoughts maybe garbled. But it would be interesting to see your take on how to use ChatGPT, without getting caught up in its coils. I note that using this has not significantly altered your personal style, which is still direct, punchy, and deals with basics, though I suspect it takes a lot of effort to make it that way.
I am quite sure I remain in control of ChatGPT, but it has had a lot of material to use to learn how I write. It has been trained on 24,000 blog posts plus a few books – so maybe 8 million words.
I want to preach to the unconverted. I asked ChatGPT: Consider the Conservative party of the UK. Given its history, policies and manifesto, how might it reframe its programme in the context of modern monetary theory?
Part of the response:
MMT as pragmatic patriotism, not radicalism
Embrace the logic of MMT while staying ideologically consistent, framing themselves as the only party disciplined enough to apply it responsibly in contrast to Labour as “reckless spenders”:
From “fiscal discipline” to “inflation discipline”
Reframe credibility not around balanced budgets but around credible inflation management as the partymost trusted to stop government profligacy tipping into inflationary chaos.
Taxation as behavioural shaping, not revenue raising
Push tax cuts for business as “supply-side inflation control” (encouraging productivity and investment). Justify taxes, on e.g. carbon, vaping or gambling as non-fiscal tools to influence behaviour.
Debt as national savings, not burden
Frame gilt issuance as a service to savers, pension funds, and global investors, consistent with a Conservative defence of strong institutions and the City of London.
Public investment justified by national interest
The UK has idle resources, and spending on big infrastructure increases security and competitiveness without needing to “find the money.”
“Sound money Conservatism” redefined
Instead of presenting deficits as moral failings, Conservatives could present themselves as guardians of inflation stability and defenders of sterling’s credibility in global markets.
ChatGPT offered a draft of a Conservative leader’s speech, which began…
For too long, our politics has been tied to a household analogy. We have been told that government must balance its books like a family balances a budget. But Britain does not run a household. Britain issues its own currency.
That means we will never run out of pounds. The real limit to what we can do is not the Treasury’s bank balance, but the resources of our great nation — the skills of our people, the energy we can produce, the technologies we can develop.
…and ended:
This is the new era of responsible Conservatism. Not bound by outdated myths about running out of money. Not reckless with our currency power. But firm, disciplined, and ambitious — for Britain.
We alone will ensure that the pound in your pocket remains strong, while your country invests in the future.
Friends, the world is changing. And with this new Conservative vision, Britain will not only keep up — we will lead.
In a sense slightly worrying, but it will never happen, I think I can confidently say.
Apologies for the visual horror of my post. I hadn’t realised that paragraphs would emerge merged.
They don’t. That’s just how the preview shows. It’s annoying and there is nothing I can do about it.
yes.
I pose questions to AI but I suspect I’m scratching the surface