I’ve been doing AI Transformation in one form or another since early 2023 – it was never my core role, but as a former CTO it’s often ended up being a substantial part of my input to different orgs I work with/in. I have built a loose playbook of techniques that work well with AI transformation, and I wanted to see what formal frameworks exist – could I adopt one of them instead? e.g. When Agile was new, things like the Agile Manifesto and Scrum let us stand on the shoulders of giants and take the best of what we’d figured out ourselves, combined with the wider/deeper ideas of industry leaders.
I did a quick review of the top hits in this space (what I’d like to do next: dig into what some of the people I follow in AI circles have said on this – e.g. Ethan Mollick … and possibly publish my own, since it’s a lot better than most of what’s out there today)
AI Transformation Review
Approach: search google for “ai transformation framework”
1st hit: McKinsey: “How to implement an AI and digital transformation”
Link: https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/rewired-to-outcompete
Click through to the link: It has nothing to do with AI. Literally: “AI” is missing from the actual page title, it was only an SEO hack that misreported the title to Google to drive traffic.
AI is mentioned a few times in the article as a copy/paste where they changed “digital” to “digital and AI”. Looks like a junior consultant cranked this out (or maybe a marketing employee?) with the minimal possible effort.
Every time I think McKinsey & Co can’t embarass themselves any more than they already have – with their appalling lack of knowledge/understanding of AI – they prove me wrong.
One thing is for sure: McKinsey has way too much money to spare if they’re burning $$$ to pay for the top Google hit on a topic they know nothing about and don’t take seriously (other than “it’s just another keyword to jump on and sell ourselves”)
2nd hit: (ScienceDirect SEO): Swedish Centers for Digital Innovation
Link: https://www.sciencedirect.com/science/article/abs/pii/S0007681325000023
This is Elsevier’s SEO promoting an academic paper. The abstract reveals it is indeed about an “AI Transformation Framework”.
How do you tell me you haven’t used AI yourself in the workplace, without saying you haven’t used AI in the workplace? You talk about “automation” as the primary role of AI. Yep: “three dimensions critical to successful AI transformation: automation, augmentation, and data richness”.
Half their citations are from 2020. The rest are from 2024. Given how long it takes to study, write, and publish, this strongly implies their paper has nothing to do with AI Transformation as we mean it in a modern (post 2021) sense: their work predates ChatGPT.
It’s unsurprising therefore that this paper appears to have nothing to do with AI beyond applying existing academic basic approaches to strategy.
Note: ChatGPT itself will create exactly the same content for you if you ask it nicely; all they’ve done in the abstract is describe an extremely common generic business process.
Note: Since this is Elsevier, the article has to be purchased – so I didn’t read the full paper.
3rd hit: Accenture: “The art of AI maturity”
Link: https://www.accenture.com/gb-en/insights/artificial-intelligence/ai-maturity-and-transformation
Starts with “In 2021…” so we already know it’s not about modern AI at all. They specifically mention lots of applications all of which are either Computer Vision or Machine Learning – none of which is particularly relevant today, they’ve all been eclipsed by any 12 year old child with access to ChatGPT.
They go on to quote statistics generated by “our machine learning models” – ouch. So they’re relying on ML predictions/guesses (which are generally inferior to LLMs) rather than on actual analysis and statistics.
Their dataset is limited to “2010 to 2021” – again: completely missing-out on modern AI.
The entire page is about pre-AI. While it’s pretty and has lots of visual polish, the page was mostly irrelevant in 2023/2024, let alone 2025. But this is what they’re paying tens of thousands of dollars (or more?) to promote on Google.
4th hit: “7 AI Transformation Strategies in 2025”
Link: https://research.aimultiple.com/ai-transformation/
Published: 14 apr 2025. Explicitly mentions “in 2025”. Could this be the first hit that actually contains an AI Transformation Strategy?
Yes. But don’t get excited yet.
It’s explicitly about modern AI! But … the author comes out swinging with this erroneous rigid belief “Digital transformation is required before companies can start their AI transformation because digital data is necessary for AI training and digital processes are necessary to roll-out AI solutions in most cases”
No, no, and no.
Further: the author appears not to have used the examples he cites, since some of them famously still do not work commercially (as of 2025, when he published this). Some work academically but are vastly too expensive to deploy to production, others slow down the organization by a year or more for only marginal benefit.
Ultimately I would recommend NOT doing most of the things on this page – they are very much on-topic, but they are a poor strategy choice (at least today) and will cost you time, money, and mental energy you could have used on something successful instead.
But the page is well-written with frequent citation links (a lot more than the others have done, where they only speak in vague generalities) — and maybe next year these steps will actually work in real companies.
5th hit: “AI Transformation Playbook: The Ultimate Guide”
Link: https://www.scalefocus.com/blog/ai-transformation-playbook-the-ultimate-guide
It’s about AI and transformation – great! It’s not wrong, but it’s not right either.
This is not the playbook it pretends to be, and it merges a lot of different ideas from different frameworks/playbooks, creating a bit of an unguided mess overall.
Ultimately it has little content and is a thinly-veiled advert for the author’s “AI Adoption Framework” (which itself is not a framework, it’s a process for you to pay them money to act as consultants. It’s literally a list of things they want you to pay them to do, not a list of things you need to do or assess).
Fine but not very helpful.