Blog
Byte-sized AI: Eleven at once
I’m back for another Byte-sized AI issue, and as promised here are my three things: 😁
- Something that you can do to improve how you work
- Something techy from the news
- Something that made me laugh

📰 The one item of news:
Eleven at once
This is not industry news as such, but it’s too big for the tip section and too significant to leave out.
I’ve spent this morning doing something that would have sounded ridiculous a year ago. I’ve had 11 AI sessions running at the same time, each one getting on with its own bit of the work!
The number’s not really the point. What took the effort was stopping the AI sessions tripping over each other. So I asked the AI to plan the work properly: break the project into tasks, work out which tasks would need the same piece of information, then group them into “waves” – sets of tasks that can safely run at the same time. Finish wave one, start wave two, and so on. It did the planning, and then a room full of the same AI carried out its own plan.
Why does this matter to you?
Because you’ve almost certainly got a job that looks exactly like this.
Take a quarterly board pack – finance narrative, sales update, ops update, risk register, exec summary. Five pieces of writing, five Copilot chats, all going at once. The risks are the same. Finance and sales both need a revenue figure, and if they each go and find their own, you get two numbers and a very awkward board meeting. Plus, the exec summary can’t be created until all the other sections are complete.
You don’t have to work everything out yourself though. Paste the list of sections and file sources into Copilot and ask:
“If I gave these to five people to write at the same time, what would go wrong?”
It’ll spot the problems. Then ask it to sequence the work so nothing conflicts. Two prompts, and you’ve got the same method I used.
You can see a visual example of what I mean below:
Future Workplace
Running several AI sessions at once without them tripping over each other
A quarterly board pack, written by five Copilot sessions in parallel. The work is split into waves. Sessions in the same wave don’t share a source, and each wave only starts once the previous one has finished.
Three sections that don’t depend on each other. Each has its own source; the brief fixes the one they share.
WaitingNeeds the finished finance and ops sections, so it has to wait for wave 1.
WaitingThe summary draws on all four sections. A final session then checks the pack for inconsistencies.
WaitingThere is a catch.
Somebody still has to read five documents and decide whether they tell one story - and that somebody is you 🫵. It's less like getting an assistant and more like suddenly managing a small team, which is a different job and not everyone's idea of a good time. But it's the direction things are going, and I'd rather you heard it from me than found out the hard way.
😂 The one thing that made me laugh:
Something that made laugh
Or at the very least go hmmmm.
Remember the things you learned at school and college? And how they apply to your current role?
At no point in my 80s and 90s schooling did anyone teach me the skills to discuss “we should run this through a deterministic wizard before we release it to AI”. In fact, I’m pretty sure no-one mentioned wizards, deterministic or otherwise, at all.
💡 The one thing you can do next week:
Forgetful Colpilot
Word can make a really good scratchpad (think of it as a shared notebook) for Copilot. Have Copilot do what you need, and have a file in SharePoint or OneDrive called copilot-scratchpad.docx (or whatever you fancy). In the prompt tell it to store a report of what happened in that file, then in future prompts include “check /copilot-scratchpad.docs”.
This gets over the problem of Copilot forgetting what it did last week.
If you are interested to find out how AI, automation and all things clever can help your organisation why not take a look at Future Workplace, our super special IT consultancy service that can supercharge productivity, business efficiency and so much more.

Subscribe to Byte-sized AI
Check out past Byte-sized AI issues:
- Byte-sized AI: Eleven at onceBuild and manage your own AI team and how to make Copilot less forgetful!
- Byte-sized AI: Do you like to read?Find out how Copilot is evolving, and make sure that you give it a job. Plus, some great reading recommendations.
- Byte-sized AI: Watching Paint DryThe rise of Small Language Models, have you watched AI Confidential and build you own Microsoft Copilot Agent.
- Byte-sized AI: Standing-on-towelCyberScotland Week, standing-on-towel and Authenticator.
- Byte-sized AI: Bubble, bubbleIs the AI bubble about to burst? Using Copilot in Microsoft Outlook. Ethics and morals, the big difference answered.
- Byte-sized AI: Did you miss me? I bet you did!AI adoption in UK SMEs, how amazing House is and the amazing ever evolving Microsoft Copilot.
- Byte-sized AI: AI overviews and using personasAI in search is here, some of it great, some not! 😂 Plus, make your AI tools ten times better.
- Byte-sized AI: Numbers tell a storyDevelop your team for AI, Copilot for Notebook and pining for the fjords!
- Byte-sized AI: Lost in an open plan area!Find out what Nick has been up to at the Kent and Medway Business Summit and have you tried triple promoting.
- Byte-sized AI: AI actFind out how AI is being kept in check, decide if AI is funny, and have you started going loopy!
- Byte-sized AI: AI opportunities£7 million, belly laughs and AI distilled! What more can you want? Byte-sized AI from the mind of Nick Ellis!
- Byte-sized AI: Are you thinking deeper?I'm back for another Byte-sized AI issue, and as promised here are my three things. 😁
- Byte-sized AI: Its a little bit of snow!This week we have seen the government get behind AI in a big way. This was inevitable whoever was in power, I think. There are a few obvious challenges here:
- Byte-sized AI: Say hello to ClaudeFor the more technical parts of my job, I use an AI called Claude, which was created by Anthropic.
- Byte-sized AI: Granny is here to make it all better!O2 have created an AI that sounds like your gran! The obvious question is “why?”. Firstly, it doesn’t sound like your gran specifically, that would be weird; it’s a generic gran called Daisy, complete with confusion about technology and a tendency to talk about her cat too much. Still doesn’t answer why though.
- Byte-sized AI: A llama with AI and an admission!Meta are doing something very interesting. They have a set of AI models collectively called “Llama” which anyone can download and use in their products for free.
- Byte-sized AI: Something special is happening in the UK!The UK is doing something a bit special. The global average for businesses adopting AI to some extent is around 13%.
- Byte-sized AI: Up, up and… back down! 😕Every so often in the world of tech something happens that is so astonishing, so utterly gob-smacking that we should all just stop and applaud.
- Byte-sized AI: An extra brain in the room!I'm back for another Byte-sized AI issue, and as promised here are my three things! 😁
- Byte-sized AI: The machines have taken over SainsburysWelcome to the very first issue of my new technology newsletter Byte-sized AI. Each issue I will give you three things: Something techy from the news Something that made me laugh Something that you can do to improve how you work
- Byte-sized AI: Be the first to hear from NickWant to get the latest news about technology and AI, then there is only one person you need to hear from and that's me!!

Nick Ellis
AI and Innovation Lead
Nick Ellis is the AI and Innovation Lead at Select Technology. He is the driving force behind our Future Workplace service, which enables SMEs across all industries to use AI and automation to enhance business operations
Throughout his career he has been instrumental in helping organisations across London and the South East, to unlock the power of AI and automation to build smarter workflows that remove manual administration and connect systems. He has helped organisations replace inefficient tools with simplified processes, freeing up teams to focus on what they do best and enabling business growth without increasing costs.
With over 30 years’ experience in business technology, spanning FTSE 500 organisations and SMEs, Nick specialises in Microsoft 365, cloud software, AI solutions, and business intelligence.