⭕️ Most AI adoption fails not because the tech doesn’t work, but because there’s no clear business case. Harnessing AI helps you create policies, training, ROI measures, and governance so AI delivers value instead of chaos. amazon.co.uk/stores/Rob-May/…

AI governance doesn't have to mean slowing everything down. This practical six-layer framework covers everything from AI inventory and responsible deployment through to human oversight and compliance, giving you a clear map of where your organisation stands and where the gaps are. I’ve explained it in more detail on my blog: thoughtprovoked.co.uk/blog/s…

⭕️ Did a French-speaking robot called PMF start my fascination with AI? thoughtprovoked.co.uk/blog/d…

More model choice coming to Copilot. Welcome Grok!

Official Trailer - iPhone Duo

What did the task you just automated used to teach the person doing it? thoughtprovoked.co.uk/blog/w…

Don't let unapproved AI use cast a shadow over your organisation’s security. Understanding employees' needs with AI tools will help ensure they can access the benefits securely. Read our new blog on shadow AI here ⬇️ ncsc.gov.uk/blogs/the-hidden…

Less prompting. More delegating. OpenAI's GPT-6 Astra is rolling out in Copilot Cowork and Copilot Studio—expanding model choice and what you can hand off to Copilot. Grounded by Work IQ in your files, meetings, chats, and business data, within your existing permissions, Copilot brings frontier model reasoning into AI built for work.

⭕️ OpenAl Astra (GPT 6): the hype, the risk, and the bit nobody's actually verified yet. @OpenAI's new model can find and exploit security flaws on its own. Its president says it might be AGI. Most businesses won't touch it this year. Consider this. The capability inside a frontier model today is the capability inside your everyday tools within eighteen months. Copilot, your CRM, next year's finance system update, all of it inherits from this generation. l've written a piece on what leadership readiness actually looks like before that capability lands, not panic, not dismissal, but three genuine questions about ownership, security assumptions and the difference between delegation and abdication. The capability curve isn't waiting for anyone's governance to catch up. The only real choice is whether you build that readiness now, while it's still your decision to make. You can read my blog here: thoughtprovoked.co.uk/blog/o…

⭕ I've just added a new section to the Thought Provoked website called Thinking out loud, in public. Most of what I write ends up in a book, in one of my masterclasses, or in a keynote, which means it's usually been through several rounds of thinking before anyone sees it. This is the opposite. It's the earlier stuff, the ideas still forming, the practical questions that come out of a client conversation or a room full of leaders asking what AI actually means for them. You can expect commentary on AI, cybersecurity and cyber resilience, shaped directly by the questions I get asked and the challenges I see inside the organisations we work with. Some of it will be fully formed. Some of it won't be, and that's the point. If you want a more honest, in-progress look at how I'm thinking about this stuff, you'll find it here: thoughtprovoked.co.uk/blog Have a look and tell me what you think, I'd genuinely like to know. You can also sign up to automatically receive these posts in your inbox if that's of interest.

Explore the latest features and updates for Microsoft Copilot in the August What's New in Copilot blog! msft.it/6017artyL

Let me tell you a story... thoughtprovoked.co.uk/blog/l…

OpenClaw 2.0 has arrived openclaw.ai/blog/openclaw-2-…

⭕️ Scout is Microsoft’s clearest signal yet that AI is moving from simply answering questions to doing the work itself. It’s Microsoft’s new desktop agent launched through the Frontier preview programme, it doesn’t just sit in a chat window. It reads and writes files, runs commands, works across Teams, Outlook, OneDrive and SharePoint, and takes action on your behalf rather than waiting for you to ask. What caught my attention wasn’t the capability, it was the access model. Getting Scout running takes two separate admin gates, an Intune policy, a formal opt-in covering third party data routing, and a dedicated licence on top of your usual Microsoft 365 setup. Sensitive actions still need your approval before anything sends, shares or updates. That’s exactly the balance I keep coming back to with clients. Govern what matters, don’t slow what works. Scout is built with real guardrails, not because Microsoft doesn’t trust the technology, but because handing an AI agent the keys to your files, your inbox and your calendar is a genuinely different risk profile to a chatbot that only talks. If you’re an SME leader watching this from the sidelines, the question isn’t whether agentic AI arrives in your business. It’s whether your governance catches up before it does. I’ve written a more in-depth explanation and and answered more questions in my latest blog post which you’ll find here: thoughtprovoked.co.uk/blog/m…

big

Why do IT support companies keep blocking the Claude AI Microsoft 365 connector over data security fears? It's one of the questions I frequently get asked when I'm speaking on AI. The instinct is understandable, but it's usually not the full picture. The connector uses delegated permissions, so Claude only sees what the signed in user could already see anyway. It's read only, it can't create, edit or delete anything, and every request gets logged in your normal Microsoft 365 audit trail. That's a very different risk profile from the one most people picture when they hear "AI reading our email." Here's what I don't think gets said enough though. A lot of general IT support companies simply don't have the depth of knowledge in AI and security to make an informed call either way, so the safest thing to say is no. That's not a criticism, it's an honest description of where most MSPs are right now. Saying no by default isn't governance, it's an absence of it. Real governance means understanding a tool well enough to scope it properly, checking your Data Processing Addendum against UK GDPR, confirming your tenant has actually opted in, and running a small pilot before deciding on wider rollout. That takes genuine AI security expertise, which is exactly why it's becoming a specialism in its own right rather than something bolted onto general IT support. I’ve written about this in more detail on my Thought Provoked blog which you can read here: thoughtprovoked.co.uk/blog/s… If your IT support company just blocked an AI connector, that's not the end of the conversation. It's worth asking whether they've genuinely assessed the risk, or whether "no" was simply the safest answer they could give without deeper AI security knowledge. What's actually stopped your organisation switching on an AI tool, the tool itself, or not knowing enough about it to trust it?

⭕ I recently shared something a little different. A children's storybook version of how @ramsac_ltd approaches AI, from the first conversation through to the tools running quietly in the background months later. The post didn't generate hundreds of comments. But it did something I find more interesting. Several people messaged me privately, or brought it up in conversation, to say it was the clearest explanation of what we actually do that they'd come across. That tells me the format worked, even if it didn't go viral. So I've taken the same story and turned it into a short video with sound. Same characters, same journey, a little more life to it. If you lead a business and you're trying to work out where AI fits, what good implementation actually involves, and what it looks like when someone stays alongside you rather than just switching something on, this is worth two minutes of your time.

Your AI agent is not the data controller. You are. - Resultsense resultsense.com/insights/202… Cc: @BillMew @robmay70 @NigelTozer @ChuckDBrooks @m49D4ch3lly @DT @archonsec @clarinette01 @JBarbosaPR @AshokNellikar @MHcommunicate @Khulood_Almani @Miriam_Brown979