Subscribe
Share
Share
Embed
The AI Briefing is your 5-minute daily intelligence report on AI in the workplace. Designed for busy corporate leaders, we distill the latest news, emerging agentic tools, and strategic insights into a quick, actionable briefing. No fluff, no jargon overload—just the AI knowledge you need to lead confidently in an automated world.
Welcome to the first AI briefing of 2026
from the not very luxurious location of Luton Airport as
I go to Poland for a month to go and help a
client deal with some delivery
of software and surprisingly I was mulling over what to discuss today
and I thought the obvious one to start the year is
that of course there are many many different AI models these days
and platforms with which you can leverage them don't forget that
not every platform was created equal not
every model was created equal and they're all there to serve different
purposes for example if you use Copilot
inside of your organization it's great it's embedded into office
it's embedded into windows you can ask it a trillion questions you
can try and dig into how to leverage a
copilot to deal with your mundane day-to-day tasks and software
email delivery document analysis all those types of things it's great
for stuff like that but it's also not necessarily great for I
don't know programming for example where you've got different models
that do better jobs for example you have Claude in
their opus models you've got GPT-5 codecs
and you've got the stuff coming out of Google with Gemini all
those models are created with a different
purpose in mind and they're there to do different things so just
because you've deployed one piece of software inside of your environment
doesn't necessarily mean that it's the one that's suited for every task
whilst the models are often multi-purpose and you can ask I
know Claude something about linguistics and I'm sure it will tell you
but it doesn't necessarily mean it's the best linguistics model out there
nor is Claude necessarily the best thing to send them receive emails
maybe maybe copilot is and so just bear that in mind when
you're deploying stuff test the different environments test different scenarios different test
different use cases because you're finding that as models evolve as
software improves the use cases will change and
the requirements will be different there
you go quick and easy today in the blazing sunshine before I
go and hop on the plane yeah make sure you test different
engines different models for different purposes and don't just rely on one
that one size fits all across your environment I hope that has
been useful stay tuned for more stuff coming over the course of
the next few months as we delve more into AI use
cases of AI and how to leverage your best in the workplace
I'm off to catch your flight bye for now and I'll see
you tomorrow