I spent much of the weekend redesigning my website… with the help of AI.

When I say “with the help of AI”, I don’t mean I pressing a button, making a cup of tea and coming back to a shiny new website. No sadly, we are not quite there yet. More like, it took me a day and a half of “move that”, “no, not there”, “make it smaller”, “that doesn’t work”, “put it back”… frustrating, but still faster than cranking by hand, click or keystroke.

So yes, AI did much of the coding and there is now much more JavaScript running behind the site than before… hopefully with some better functionality.

And I feel this is still a big misconception at the moment. AI is often portrayed as cheating, lazy or somehow avoiding doing work… the work is still there, it is just changing.

Years trying to keep the bots out

In this move to JavaScript, it did make me think about data, particularly search engine optimisation.

Data, your website data, is an asset and exposing it to search engines gets you found… today gets you tagged and included in AI queries.

The irony is businesses (especially large businesses) have spent an enormous amount of time over the years protecting their websites, especially from Bots. We stopped exposing data, RSS feeds disappeared and website became increasingly dynamic. All our information was safely tucked behind JavaScript, APIs, logins and various technical barriers, for human eyes only.

There were good reasons for this. Nobody wanted competitors automatically scraping prices, products and offers. As a human, you could still access the data, but as a bot, it was restricted.

All of this worked well, driving traffic to websites… but with AI that is now changing. With AI, we may desperately need and need to enable this very capability.

Something fundamental is changing

Gradually we changing how we consume online information. The days of surfing are a faded past, the rising light of search is setting and social media is getting swamped… now? We just ask a question to AI.

I mean, why visit ten different websites to compare what each says, when I can ask one question and get the results… speed to answer is shortened.

But, this information has to come from somewhere. And, it is the bot that scrapes the data and feeds the model ecosystem to generate the response. If you want to be in that response at all, you need, build, publish and expose your data, or at least think about it and have a strategy. It may increasingly be the thing standing between us and our (human) customers.

LinkedIn lives

At work, many of us, especially if you are in marketing or sales, live on LinkedIn. Hours are spent scrolling, posting and monitoring for engagement, it has become a primary channel of communication, especially for new or industry related contacts. For many businesses LinkedIn is the only show in town.

Plenty of ink has been spent writing and thinking about optimising this, it is easy to get obsessed. When is best to post? What content works best? Should there be an image.. or a video? Is Tuesday at 8:07am somehow better than Wednesday at 10:15am?… and when all else fails… post volume.

Which is what we now see… huge volume of posts… often AI ‘enhanced’ (yes, some of it is mine!)… and as a result, on the platform, as a user it is sometime it feels hard to find the content you are looking for, in all the noise competing for attention.

Curious… and wondering why ‘such great content such as this’ 🤣 does not get the traction it should (in my mind 😆)… I spent some time looking at my own feed, to understand what I was actually going on and it was revealing.

• Nearly half of what appears in your feed, isn’t really organic content from people I know – it is paid for or LinkedIn suggested content.
• 10% of the posts get 75% of the attention.
• Many posts are weeks old, bubbling with comments rather than something new.
• New posts, don’t really appear. Yes you can look at ‘recent’ posts rather than new… but the volume of content is huge, unmanageable.

So it is easy to publish what you think is a really interesting piece of content and then almost nobody sees it…

We talk about “posting on LinkedIn” as though we are broadcasting to our network. The data shows… this isn’t the case, getting noticed is much harder than we may think.

LinkedIn is social media now

I know we know this, but the parallels with other social media platforms were strong. Deeply researched, professional, insight-heavy material, does not perform well, unfortunately. Despite LinkedIn’s branding as a professional platform, it seems, well we are just not really that interested in it.

Content that did get noticed, and engagement was the often personal, sometimes opinionated and slightly edgy posts. These are the thinking people react to… and especially comment on. It gets seen.

For those of us who remember the old, professional network, corporate publishing platform, I can feel your heart sink. LinkedIn is undoubtedly a social network now, optimised for interaction.

This is not just an opinion, it was the data said, and I suppose something we need to adapt to.

Rethinking digital

And, that is a theme to think about this week back in the office… adaption to this new digital reality that is developing in front of us. What do we really need to do with our data, how will it be used and what do we need it to be used for.

The world of digital is clearly changing… we need to adapt.

Have a good weekend everyone

The last few months, I have spent plenty of time using AI to build applications, automate tasks and generally see what is possible… not leading edge, just helping out.

The experience has been enlightening and humbling.

What would previously have taken weeks can now be created in days… sometimes in an afternoon. It is like having a development team inside your laptop.

… and the faster and easier it is build… the more exciting it is… but, the less time you get to focus on what is really going on underneath.

Obviously this results in risk (and a whitepaper, more considered that my blog is a here), but here are some that really stood out

What did I just approve exactly?

AI is no longer just providing answers. It now runs code, installs software, opens files, calls APIs and can make changes on your computer.

That little approval box appears, ‘do you approve’. We click yes. Everything carries on.

But did we check the code, do we really understand what is it doing?

Probably not… honestly… of course not.

I mean do you want to spend 3 months coding, or bash an output out this afternoon to test the idea!

It is a risk, one that human psychology is making it hard to defend against.

Who else is in the room?

Most AI processes are not powered by just one system, they are models, an ecosystem of APIs, installed packages and skills.

It all feels seamless, but underneath what is happening to the data, which firms are touching the data, where, how? … and we are increasingly reliant on these models… what happens if we lose access, or in my case, the price goes up?

We are quickly stumbling further down the rabbit hole of extended (data) supply chains. Whilst this probably doesn’t matter much when we are building small, JavaScript, stand alone websites… okay I put the wrong bins out on Wednesday…. but, beyond this, into enterprise processes it is something we need to understand for sure, and fast.

A numbing effect?

I hear… ‘we managed the AI risk, by putting a human into the process’. Human oversight, human decision.

As a human, this is all very reassuring… but six months later you can foresee the problem.

After six months of approving nearly every decision, humans get bored… checks become cursory, a rubber stamp… and then the mistakes so to filter through (small in number but maybe consequential).

… how do we think about “the computer said yes… but the human says no” to customers. Something to consider.

The new legacy process?

This AI rollout is increasingly reminding me of the early days of spreadsheets.

Anyone could build something useful… nobody documented it.. then the person who built it and actually understood left. The spreadsheet continued running… for the next 15 years, and no one dared to go near it.

This pattern could easily repeat… but as it is AI… faster.

We have redesign and rebuild processes in an afternoon. But who owns it afterwards? Who supports it? Does the next person understand why it was built that way? Can we may changes?

Today’s automation may very quickly become tomorrows legacy system. A spaghetti diagram of legacy systems, connected by AI pipes, that no-one understands… locking us into vendors and providers long term. Something to avoid.

Downskilling?

You can already feel it. AI is changing skill-sets. Tasks, often mundane, I once completed manually are now delegated.. to my AI assistant. Yes, it has saved time… but also means I practice less often.

The same thing has already happened with my handwriting… I only really now type and my handwritten notes are, as a result, only remotely legible.

Of course new skills are coming in too, but staying up to date, and picking where to continue to practice existing skills (such as writing) is going to be important.

Lastly… sameness

It is me, or with the explosion in the use of AI, is everything is starting to look the same?

The same style. The same language. The same graphical format.

I know this is something I am guilty of too… everything looks really polished… but the question I always have is… does it stand out, especially when AI content is becoming increasingly easy to spot.

Being different matters, and if we are all creating content from the same handful of models and tools, it can relatively easily all look the same… no matter how good the content, it will not stand out.

Being human, we know what being human actually feels like… what is funny, on trend, irrational and those slightly odd ideas that do connect.

And, this is something we need to remember, in this blur of AI development, back in the office. For connecting with real customers… our humanness is in some ways our advantage.

READ THE WHITEPAPER HERE

For those of us bottom feeding on cheap(er) AI services, we got a shock this week. DeepSeek increase the pricing of their model APIs in some cases by 14x. This means, of course, bills of $5 a month jump to $25, and $25 to $125, with no real change in functionality, just a straight increase. And, this is on the back of the peak time pricing, introduced a few weeks ago too.

Now I am not begrudging DeepSeek an increase their prices. The price was already low and by all accounts they do seem to have been struggling with capacity (and also looking for funding, so more revenue no doubt helps).

Even with the price increase they are still amongst the cheapest in the market.

But, this is happening elsewhere too. X.AI, in May this year, pushed users away from their extremely well priced “Fast” models to Grok 4.3, a 5x increase. It all points to a wider trend around AI pricing… it is going up.

The API economy

In our daily lives, most of us are using AI models now. If nothing more it is a much more efficient way to search for information (is it better or more accurate, that we can argue about another time). However if you are mainly using the browser or app based chat interfaces, with bundled pricing, you may be unaware of the burgeoning economy and infrastructure that is being built underneath. I know I was.

Beneath the surface of AI services is typically a myriad of APIs and API calls. These APIs are simply agreed data exchange formats that service can use to exchange information, to call out to other programmes and services. Some you pay for, some are free, but they are intertwined in modern AI infrastructure.

LLM Model, Web-search, Image creation, Email, Calendar, Weather, Daily Joke of the day… all can be linked to APIs… even the FCA is getting in on the act with the release of an API to access their Handbook.

Each of these have price points, data security considerations and with this complexity comes a requirement for some sort of active management.

• Do I need to be using the most expensive and capable model for simply moving a document?
• Do I use a particular image model to create a visual summary?
• Which gives me more accurate and current, up to date, information when searching?
• What data am I exchanging and where is it going?

Of course, it may be worth paying the price for a premium model, that does all this for you. I feel sure many will, and large providers are adding functionality at pace (eg GrokBot, by SpaceX launched this week… a bit like OpenClaw (now owned by OpenAI)).

However for those of us without oodles of cash to burn (ie many businesses and certainly me!) we will be left managing this complexity. Managing it well can have a dramatic difference to the cost profile (I was able to reduce cost by 80% with some optimisation).

Yet if monthly costs are now going to routinely reach £40 a month (or even £100+… I seem to be spending even more when you add in all AI services) per user, it does feel like the sun is setting on the era of cheap AI… winter is coming and we need to be prepared!

Going Local

So it is at this post we should ask the question… when do we move away from server based models to highly capable local, open source, models on our own, local, server.

These models are competitive, advanced and this is already possible (Ollama is super easy, and easiest, to try if you have not). However I have found without very a high end PC, memory and graphics card, running the most capable models are still slow in comparison to online versions.(ish)… and for higher end services such as web-search, image and video creation, in some cases not possible on my existing (dedicated) hardware.

This is really only a question of processing power, typical consumer and business PCs are just not fast enough. To build or buy, one that is currently costs around £5,000. Pricey, yes… but vs a steeping monthly AI bill… it is getting there for the small user.

I wonder, with the likely price escalation, as firms recover investment costs, if this is a trend will see as people react and adjust.

Nothing new?

I suppose in some ways this is nothing new.

Already in business we use different employees with different skill-sets (and price points) for different tasks. We match the resource to the task to optimise the cost profile, we give them the tools to communicate and measure performance for jobs done well… and we also look at building in-house, locally, vs using third party managed services (or not).

And, this returns to a recurring theme.

Despite all the excitement around new tech and AI, in some ways not much has changed. Much of the toolset we need to manage it we already have…

AI is not making good management, structure and process go away, in many ways it is making it even more important.

Something to think about back in the office this week. Have a good week, everyone.

[A quick event plug. We will be talking about many of these issues, including governance, standards and risks at our live events in Oct/Nov… more info here, come along if you are free -> Event page]