A conversation I’m having more often with business owners goes something like this: “We don’t really use AI yet.”

A few questions later, it turns out somebody is using ChatGPT to help write emails.

Someone else is summarising meeting notes with an AI tool.

Another person has found a browser extension that helps them create content faster.

The business may not have formally adopted AI, but AI has arrived anyway.

That’s not surprising.

When people find a tool that helps them get through their work more quickly, they’re going to be curious about it. If it saves time, they’ll keep using it.

The difficulty comes when nobody has stopped to agree where the boundaries are.

Imagine a member of your team receives a long customer email and pastes it into a public AI tool to help draft a reply.

The response comes back in seconds and saves them fifteen minutes of work.

Tomorrow they do it again.

A week later they upload a proposal because they want a summary.

A month later somebody else is using a different AI tool to analyse a spreadsheet.

At no point does anybody feel they’re doing something risky. They’re trying to work more efficiently.

Yet customer information, financial data, internal documents, and business plans can end up being shared with systems that haven’t been reviewed or approved by the business.

That can happen without any bad intentions.

I’ve also noticed that AI spreads differently from other technologies.

A new accounting package usually goes through a buying process. A new CRM gets discussed before it arrives. But AI tools are discovered by individuals during the working day and adopted immediately.

By the time a business starts thinking about an AI policy, people may already have their favourite tools and workflows.

That creates an awkward situation.

If somebody believes a tool is helping them do their job better, they may not understand why they’re being told not to use it.

Quite often, they feel they’re solving a problem rather than creating one.

So, to get the most value from AI, have open conversations about it.

Make sure your team all understand which tools are being used, where information is going, and what data should never be shared outside approved systems.

That gives people enough freedom to benefit from the technology without accidentally creating risks for the business.

If you’ve never asked your team which AI tools they’re using, the answers might surprise you.

It’s a good place to start.

If you’d like help reviewing how AI is being used across your business, and making sure it’s helping rather than creating risk, get in touch.

AI was supposed to remove repetitive work.

In some businesses, though, something different is happening.

People are spending large parts of their day moving information between systems so the AI can function properly.

Copying notes from one platform into another, checking whether data matches in different apps, rewriting prompts to give an AI tool more context, and manually correcting outputs that almost worked, but not quite.

Sound familiar?

There’s a name for this: Human middleware.

Employees end up acting like the glue holding disconnected systems together.

Once you notice it, you’ll start seeing it everywhere.

Someone downloads information from one system because another can’t access it directly.

A team member pastes customer details into an AI tool to generate a response, then copies the finished result somewhere else.

Data gets checked manually because nobody fully trusts what the systems are producing automatically.

It eats away at time.

The strange thing is that businesses can still feel more productive overall while this is happening.

AI genuinely does help people move faster in many situations. Emails get drafted quicker, reports take less time, and information becomes easier to summarise.

But new layers of admin appear around it.

AI tools often arrive faster than the systems underneath them evolve. A company adds one assistant here, another automation there, a separate AI-powered feature somewhere else… but the tools don’t naturally connect in a smooth way.

So, people bridge the gaps manually.

That creates an odd working environment where employees spend increasing amounts of energy translating between systems instead of doing the work those systems were meant to support.

And that can quickly become exhausting.

You end up with busy days that feel productive on the surface, but a lot of effort is going into coordination rather than progress.

If systems don’t integrate properly, if data quality is inconsistent, or if processes still rely heavily on manual handoffs, AI can sometimes layer extra complexity on top rather than removing it.

That’s why you may need to approach it differently.

Instead of adding isolated tools everywhere, focus on how information moves through the business.

Look at where data lives, how systems connect, and whether people are still spending too much time acting as the translator between technologies.

Ultimately, your team shouldn’t be spending their day helping software talk to other software.

They should be spending their time solving problems, helping customers, making decisions, and doing the work that creates value.

If your employees are constantly switching between apps, correcting AI outputs, or manually stitching workflows together, that’s a sign the technology strategy needs tightening up a little.

Is it time you reviewed your technology to make sure it’s reducing workload, rather than creating more of it behind the scenes? We can help. Get in touch.

Most cyber security tools work reactively.

Something suspicious happens, the system spots it, and then tries to stop the damage before it spreads.

That’s incredibly important.

But Microsoft is working on something that pushes much further: Using AI to find weaknesses before attackers discover them.

The new system is called MDASH, and the idea behind it is genuinely interesting.

Microsoft has built a platform that uses more than 100 specialised AI agents working together to search for hidden security flaws inside Windows.

These agents are designed to inspect different parts of the system, test for weaknesses, and flag potential vulnerabilities automatically.

Simply put, Microsoft is using AI to hunt for security holes at a scale humans simply couldn’t manage alone.

And it appears to be working.

During testing, the system reportedly uncovered multiple previously unknown vulnerabilities inside important parts of Windows.

This included flaws that attackers could potentially have exploited remotely over the internet.

Some of those vulnerabilities were considered critical.

These are the kinds of weaknesses that, in the wrong hands, could potentially allow attackers to take control of systems or run malicious code.

What makes this more impressive is the accuracy.

One of the biggest problems with AI-driven security tools has always been false alarms.

Systems that flag hundreds of “possible issues” which turn out to be nothing. That creates noise, wastes time, and makes security teams less effective.

Microsoft claims MDASH has been unusually good at avoiding that problem while still finding genuine risks.

Now, before anyone assumes AI is about to solve cyber security completely, it’s important to keep this in perspective.

This technology is mainly being used internally by Microsoft engineers now.

It’s still early days, and even if these systems become more widely available, they won’t suddenly replace the fundamentals that keep businesses safe.

Because most cyber attacks still succeed through ordinary gaps:

• Weak passwords
• Unpatched systems
• People clicking the wrong link
• Poor access controls
• Missing backups

Those remain the biggest risks for most businesses today.

AI-driven security tools may eventually become a powerful extra layer of protection, especially for large organisations managing huge and complex systems.

The idea of intelligent agents constantly scanning for hidden weaknesses before criminals find them is a very promising direction for the future.

But the basics are more important right now.

A fully patched system with strong passwords, multi-factor authentication, good backups, and sensible user awareness training will protect most businesses far better than chasing the latest AI security trend without solid foundations underneath it.

The future of cyber security probably will involve more AI working behind the scenes, both defending systems and, unfortunately, helping attackers too.

But while the technology evolves, the core principles of staying safe haven’t changed.

Good security still comes down to reducing risk, limiting opportunities for attackers, and making sure the simple things are consistently done well.

If you want to make sure your business security is up to scratch, we’d be happy to help. Get in touch.

Are you tired of that slight pause when you open something on your PC and it takes just a fraction longer than you expect?

It’s not enough to stop you working, but it is enough to irritate you for a moment.

That’s the sort of thing Microsoft is fixing right now in Windows 11.

In recent months there’s been a focus on improving the overall feel and responsiveness of Windows, rather than simply piling in more features.

And honestly, I think it’s the right priority.

The latest work happening behind the scenes is all about making core parts of Windows faster and smoother to use.

A big part of this centres around something called “WinUI”. You can safely forget the name after reading this, but it helps to understand roughly what it is.

WinUI is the framework Microsoft uses to build the visual side of Windows apps and interface elements.

It controls how things look and behave on screen. File Explorer, menus, settings windows, and many built-in apps all rely on it.

Microsoft has been gradually modernising Windows 11 using this framework, but now it’s also trying to make the framework itself run more efficiently.

When the foundations get faster, everything built on top of them benefits too.

The company has been testing improvements using tools like File Explorer and Notepad as benchmarks. File Explorer, for example, is the part of Windows you use constantly to browse files and folders.

According to Microsoft’s engineers, they’ve managed to significantly reduce the amount of background work Windows has to do when launching these apps.

Less unnecessary processing means things should open more quickly and feel more responsive.

You probably won’t see dramatic “twice as fast” changes overnight. This is more about shaving off small delays across lots of everyday actions. Opening folders, switching windows, launching apps, etc.

Across an entire workday, these moments quickly add up.

There’s also another interesting part to this story.

Microsoft’s different Windows teams appear to be working together more closely than they used to.

That might not sound exciting, but when separate teams optimise different parts of the system in isolation, progress can feel inconsistent.

What’s happening now seems more coordinated.

And it’s improving how the whole experience fits together.

There’s also work being done on something called “low latency”, which basically means reducing the delay between you doing something and Windows responding to it.

Again, not glamorous… but noticeable when done well.

People work faster when systems feel responsive. Frustration drops, small interruptions happen less often, and everything feels a little smoother and more predictable.

And that’s what good technology should do.

If your business PCs are starting to feel sluggish, or you’re not sure whether your hardware is getting the best out of Windows 11, we can help. Get in touch.

Most businesses have a handful of phone calls that happen time and again.

“What time do you open?”

“Can I change my appointment?”

“Has my order been dispatched?”

“Can somebody call me back?”

Of course, these are all normal questions for someone who doesn’t know the business.

But when your team answer the same things all day, it eats into time that could be spent on more valuable work.

That’s why AI voice assistants are getting so much attention right now.

And no, I’m not talking about the old “Press 1 for sales, press 2 for support” systems that everybody hates.

This is something different.

Modern AI voice assistants can hold a natural conversation with a caller.

Instead of forcing people through rigid menus, they let callers explain what they need in their own words.

The system listens, understands the request, and responds accordingly.

In many cases, the caller doesn’t even realise they’re speaking to AI right away.

That’s a big step away from traditional phone automation.

Older systems often fall apart the moment somebody says something unexpected.

AI voice assistants are designed to handle more natural back-and-forth conversations. The better ones can connect directly into your business systems as well.

So instead of simply answering questions, they can do things like:

• Book appointments
• Check account information
• Update tickets
• Send confirmations

That can make a big difference for businesses handling a high volume of routine calls.

But before you rush to replace your customer service team with AI, take a moment.

This technology works best in specific situations.

If your business receives lots of repetitive, predictable enquiries, an AI voice assistant could genuinely save time and reduce pressure on your team.

Customers get answers faster, staff spend less time repeating themselves, and people are freed up to deal with the conversations that need human judgement.

Where things become harder is when conversations are emotional, complex, or unusual.

A frustrated customer with a billing dispute probably still wants a person. Someone dealing with a sensitive situation doesn’t want to feel trapped inside an automated process.

It’s a good idea to start small.

Rather than automating everything, identify one narrow area where AI can help reliably. Maybe appointment bookings. Maybe basic account queries. Maybe overflow calls outside office hours.

That approach gives you room to test what works without disrupting the wider customer experience.

There’s another side to this as well, and it’s important.

If an AI voice assistant is accessing customer data, recordings, calendars, or internal systems, security and compliance are important.

You need to know where that data is stored, who can access it, and how the system connects into your existing tools.

And not every platform handles this equally well.

Some sound impressive in demos but struggle badly in the real world when there’s background noise, strong accents, industry jargon, or unexpected questions.

Others create more work behind the scenes because they don’t integrate properly with the systems you already use.

Choosing the right setup is more important than choosing the cheapest option.

AI voice assistants will become genuinely useful over the next few years. The technology is improving quickly, and in the right environment, it can remove a lot of repetitive work.

If you’re curious whether an AI voice assistant could work in your business, we can help you explore things carefully. Get in touch.

Cyber criminals arguing with each other sounds like good news.

One group targeting another.

Threats flying back and forth.

Even promises to “help” victims caught in the middle.

It almost feels like justice.

But here’s the reality: There’s only one thing you can rely on cyber criminals for… and that’s looking after themselves.

Recently, one ransomware group started threatening another.

They claimed they would expose identities, leak data, and even offered to help victims unlock their files.

At first glance, that might sound like an opportunity. If you’ve been hit by an attack, and someone claims they can fix it, it’s tempting to consider it.

But this is where things get dangerous.

These groups aren’t operating with any sense of fairness or responsibility.

They’re not trying to do the right thing. They’re trying to gain leverage, create pressure, and ultimately make money.

Even when they turn on each other, the motivation doesn’t change. It’s still about control and profit.

In this particular case, one group claimed they could help victims recover their data.

The problem is, there’s no proof they can.

And even if they could, you’re still dealing with a criminal organisation that has no obligation to follow through on anything it promises.

It’s like being caught between two scammers and hoping one of them is the “honest” one.

That’s not a position you want your business to be in.

This situation highlights something important: When a cyber attack happens, your options should never include trusting another attacker.

It doesn’t matter how convincing the offer sounds or how desperate the situation feels. Relying on criminals introduces even more risk.

The only reliable path is through proper security measures and trusted support.

That means having the right protections in place before anything happens.

Backups that are tested and accessible.

Monitoring that helps you spot unusual activity early.

A clear plan for how to respond if something goes wrong.

And if the worst does happen, it means working with people whose job is to defend your business, not exploit it.

Cyber attacks mean making decisions under pressure. And the decisions you make in those moments can either limit the damage… or make things much worse.

If you’re not completely confident in how your business would respond to an attack, it’s time to create a strategy. If you’d like help with that, get in touch.

If you use Microsoft 365 every day, you’ve probably noticed that Copilot hasn’t stood still.

It’s been evolving in the background, gradually becoming more embedded in the way Word, Excel, Outlook and the rest of your tools work.

Microsoft tends to roll these changes out in “waves”, which is their way of grouping together bigger updates rather than releasing everything at once.

We’re now into the latest phase… and this one feels a little different.

Earlier versions of Microsoft Copilot were mostly about helping you. Drafting emails, summarising documents, creating presentations.

Useful, time-saving features that still relied on you to guide every step.

Now, the focus is shifting.

Instead of just helping with tasks, Copilot is starting to take on more responsibility within workflows.

One of the biggest changes is the ability to create your own AI agents inside Microsoft 365.

In simple terms, that means you can build small, task-focused assistants that handle specific parts of your day.

Think of something that can manage repetitive processes, follow a set of rules, and carry out actions without needing constant input.

That opens up some interesting possibilities.

You could have an agent helping with routine admin, keeping documents organised, or supporting internal processes.

Alongside that, Copilot’s chat experience has become more capable.

You can now make changes to documents, emails, spreadsheets and presentations directly from a conversation, without jumping between different screens.

There’s also a new layer being introduced called “Work IQ”. That’s Microsoft’s way of helping Copilot understand your business context better.

It uses your business’s data, communication patterns and internal content to make responses more relevant to how your team works.

In theory, that means less generic output and more tailored support.

Microsoft is also expanding the underlying AI models Copilot can use, bringing in technology from different providers.

You don’t need to worry about the technical side of that, but it does mean the system should become more flexible and capable over time.

Looking slightly ahead, there’s another feature worth keeping an eye on. Something called “Copilot coworker”, which is currently being tested.

This takes the idea of AI agents further, allowing them to carry out longer, multi-step processes rather than just responding to a single request.

So instead of asking for help with one task, you could assign something more involved and let it run.

To support all of this, Microsoft is also introducing tools to help businesses manage these agents properly.

That includes visibility over what they’re doing, how they’re being used, and how they fit into your wider systems.

Which is important, because as Copilot becomes more powerful, it needs to be properly controlled.

If you’d like help understanding how to use these changes safely and effectively, get in touch.

It sounds like a joke.

“Maybe we should just ban AI on Friday afternoons.”

But the more you think about it, the more it makes sense.

By Friday, most people are running on fumes. The week’s been busy, there are things to finish before the weekend, and attention to detail isn’t always at its sharpest.

Now add AI into that mix.

Tools like Microsoft Copilot can draft emails, summarise documents, and generate responses in seconds.

That’s incredibly useful.

But it also introduces a subtle risk.

The output looks polished and ready to send.

When you’re tired, it’s very easy to trust it without giving it a second look.

That’s where problems creep in.

The kind of small issues that slip through when no one’s paying close attention.

An email that sounds slightly off or a message that comes across the wrong way. Something that’s technically correct, but not appropriate for the situation.

The wrong tone can be harder to spot when you’re rushing to finish the week.

There’s also something else going on here that doesn’t get talked about enough.

The more we rely on AI to think for us, the less we challenge what it produces.

That can dull critical thinking.

Instead of asking, “Is this right?” the default becomes, “That looks fine.”

Across a team, over weeks and months, it can change how decisions are made.

Some people are starting to call this “AI fatigue”.

Constantly interacting with AI tools, checking them, and relying on them can feel mentally draining.

So the idea of “no AI on Fridays” perhaps isn’t so out there.

At the end of the week, people are more likely to skip steps, less likely to question things, and more likely to hit send.

And AI makes that easier.

You don’t need to switch AI off, but you do need to build in habits around it.

Encourage your team to review anything generated by AI before it goes out externally. Especially emails, customer responses, or anything that represents your business.

Make it normal to sense-check the tone as well as the content.

If you’re using tools like Microsoft Copilot across the business, it’s also worth setting some simple guidelines. Clear expectations about when AI should be used, and how its output should be handled.

AI is a great assistant, but it still needs a human in the loop.

Especially at 4:30pm on a Friday.

If you’d like help creating some sensible rules around how AI is used in your business, get in touch.

You probably don’t think twice when you see a Windows update.

A quick click, a short wait and the job’s done.

It’s one of those routine things that feels safe by default.

After all, updates are there to protect your system, not cause problems.

That’s exactly why this latest scam is catching people out.

There’s a fake Windows 11 update doing the rounds that looks almost identical to the real thing.

The page is designed to look like an official Microsoft support site, right down to the layout and language. At a glance, there’s nothing obvious that gives it away.

It presents what appears to be a normal update for Windows 11 and invites you to download it.

But click the button, and you’re not installing an update.

You’re installing malware.

Now, here’s where it gets a bit more concerning.

It isn’t a rough, badly put-together scam. The file itself is built using legitimate tools that developers use every day.

It’s packaged in a way that looks genuine, with familiar labels and properties that suggest it’s from Microsoft.

Even security software can struggle to flag it straight away, because on the surface, everything checks out.

Attacks have evolved.

In the past, fake updates were easier to spot. The designs were poor, the wording odd… something always felt off.

But now, they’re designed to blend in.

Updates often happen quickly and without much thought. Someone on your team sees a prompt, clicks to install, and carries on with their day.

That habit is normally fine.

But when attackers start mimicking trusted processes, that same habit becomes a risk.

So, what’s the safer approach?

Keep updates inside Windows.

If you’re using Windows 11, the safest place to manage updates is through the built-in Settings app. That’s where genuine updates are delivered and installed.

If you ever need to download something manually, go directly to Microsoft’s official website rather than following links from emails or unfamiliar pages.

It’s also worth reminding your team of a simple rule: If something unexpected appears asking you to install an update, pause and double-check before doing anything.

You don’t want to make your team paranoid. But they do need to recognise that even the most routine actions can be turned into an entry point if the timing and presentation are convincing enough.

The more normal something looks, the less likely people are to question it. And attackers are counting on that.

If you’d like some simple ways to reduce this kind of risk, we can help. Get in touch.