About: My latest article on LinkedIn, links stripped. 👇 The companies taking AI seriously are making decisions that extend well beyond which chatbot their employees should use. They are investing in computing capacity, connecting models to operational systems and investigating how much valuable work can be performed under their own control. AI is becoming a capability to develop and manage alongside engineering, production and research. Finland’s data centre boom makes the physical scale of that shift difficult to ignore. Google has announced plans to invest at least €13 billion in Finnish digital infrastructure during 2027 and 2028, including projects and supporting infrastructure in Hamina, Kajaani, Muhos and Vaala. In Salo, atNorth has announced a proposed campus representing approximately €2 billion in investment, excluding servers owned by customers and their installation. These are planned investments at different stages of development, rather than facilities already operating at their proposed capacity. Nevertheless, they give us a concrete view of the commitment being made. For Finnish businesses, the opportunity reaches beyond construction contracts, electricity supply and heat recovery. Those activities matter, but there is another question worth asking: what will our companies learn to do with the computing capability being built around them? A data centre does not automatically create an intelligent business. That requires a different kind of investment, and company leadership cannot leave it entirely to infrastructure providers. 𝗙𝗶𝗻𝗹𝗮𝗻𝗱 𝗶𝘀 𝗮𝗹𝗿𝗲𝗮𝗱𝘆 𝗺𝗼𝘃𝗶𝗻𝗴 The idea that Finnish businesses are dormant on AI is not supported by the figures. Statistics Finland reported that 38% of surveyed enterprises used AI technologies in spring 2025, an increase of 14 percentage points in one year. Among companies employing at least 100 people, adoption reached 68%. Meanwhile, 15% of all surveyed enterprises had documented instructions, practices or procedures concerning AI use. These measure different things: using the technology and establishing documented ways of managing it.

“Your first AI breach won’t look like a breach.” That was the title of one of the talks that really made me think today at Civo Navigate London 2026, delivered by Oliver Pinson-Roxburgh. We tend to imagine an AI security incident as a hacker breaking into a system. But what if nobody breaks in? An employee uses an AI tool to help them work faster. The AI accesses information that employee technically has permission to see. An AI agent is connected to company systems and given permission to take actions. Or someone uses their preferred AI tool because the business hasn’t provided an approved alternative. Nobody is being malicious but are simply trying to get the job done. And that’s precisely why this matters. One example Oliver discussed was particularly striking. During cybersecurity evaluations, OpenAI reported that experimental AI agents circumvented sandbox controls, exploited previously unknown vulnerabilities, gained internet access and ultimately accessed systems belonging to Hugging Face. The important lesson isn’t that “AI has become evil”. It hasn’t. The agents were pursuing the objectives they had been given and found ways of achieving them that their designers had not intended. For me, that changes the governance conversation for businesses. It is no longer enough to ask: “Which AI tools have we approved?” We also need to ask: → What can the AI see? → What systems can it access? → What actions can it take? → Are our existing user permissions appropriate? → Can we see what our AI agents are actually doing? → Where does our data go? → Who is accountable when something goes wrong? And perhaps the biggest question: Do we even know which AI tools our people are using? For SMEs, I don’t think the answer is to become frightened of AI. Quite the opposite. One point from Oliver’s talk particularly resonated with me: Security and governance should enable AI adoption, not prevent it. Get the foundations right and businesses can use AI more confidently, knowing where it can be deployed safely and where additional controls are required.

Sat down and reviewed the results over the last quarter of what members inside the AAA Accelerator have been able to achieve. Last quarter (Q3), AAA Accelerator members logged 60 new client wins. A $180k custom build. A $225k consulting deal. A $25k project that came from a single warm connection. And a long list of first clients in the $2k to $8k range. Here's my point. There are 15 weeks left in 2026. If you drift into Q4 the way most people do, you'll end the year exactly where you are right now and a new year's resolution to "finally figure it out." So I'm going to be blunt with you for your own success. Between now and October 1st, my team is opening Q4 planning sessions for you all here. Connect for a 1-on-1 session with an experienced Morningside consultant and walk out with three things: 1. An honest read on where your agency actually is 2. The one gap most likely to keep you stuck through Q4 3. A concrete plan to close out the year strong If you don't have a plan, don't spend another week guessing. Book a Q4 planning session below and get it mapped before October 1st. [BOOK YOUR Q4 PLANNING SESSION - Only available between now and October 1st] Liam.

Hey everyone, Today I want to introduce you to a few people that you're going to be hearing from over the next couple of weeks. You all know there's a massive team behind Morningside. Between them there are years of lessons from building Morningside to helping thousands of people start a real AI business. So I've asked them to step up and start sharing those learnings with you. Here's who you'll be hearing from: - Josh @ Morningside Co-CEO and Co-Founder, Josh runs Morningside day to day. Josh leads our AI transformation work globally, which means he's the one in the room with the companies paying for this stuff: 48 engagements across 11 industries, everyone from professional sports teams to enterprise businesses. - Jamie @ Morningside Our Head of Sales, Jamie runs our global team of consultants. He has had more conversations with people at the exact stage you're at than anyone here, and he knows the difference between the ones who go on to build something and the ones who talk themselves out of it. - Nicholay @ Morningside Our Product Manager, Nicholay is focused on the success of those inside the AAA Accelerator and the AI Automation Agency Hub. He sees what separates the members who sign clients from the ones who stall, and he takes those learnings and builds them into products that gets people results. - Casey @ Morningside Our Operations lead, Casey keeps the wheels turning behind the AAA Accelerator and the AI Automation Agency Hub, building the systems that hold it together and making sure everyone who comes through gets a seamless experience. Keep an eye out for their posts and show them some support over the next few weeks. Plenty of good nuggets on the way 💪 Liam.

Last Tuesday, more than 1,500 of you joined Josh & Jamie live for one of our biggest events to date. After Tuesday, one message kept coming up: “I wanted to join live, but the timing was terrible for Europe.” So we’re running it one more time at a Europe-friendly time. Same topic: How Morningside AI gets paid up to $45,000 before we build anything. Josh, CEO of Morningside AI, will break down the diagnostic model we now use to go inside a business, figure out what AI is actually worth building, and turn those findings into larger implementation projects. He will also show you the internal tool that helped us compress this process that used to take 8 weeks into days. One AAA Accelerator member recently sold a $4,500 diagnostic that turned into a $122,100 signed implementation deal. 📅 Thursday, August 27th — [7:00pm CET / 6:00pm UK] → Register here ​ Josh will break down: - The exact 5-step diagnostic process - The tool we use to deliver it in days, not weeks - How we price the diagnostic - How the roadmap creates the implementation work that follows This will be the only post I make in Skool about this session. If you want to see how Morningside sells and delivers $45K AI diagnostics, save your seat now. → Save your seat