GenAI Could Add up to $340 billion to Bank Profits....IF they get it right!
Are Mckinseys seven AI steps enough? I don't think so!
View online: here
Or download PDF
McKinsey is predicting big profits for banks using GenAI! This isn’t their first of what I like to call “big numbers” reports for AI and won’t be their last.
Mckinsey first touts big numbers then hits readers with the best sales pitch since, “Buy two get one free.” McKinsey says all it takes is “seven dimensions” and a Mckinsey consulting contract to get it right.
I think that this is a stretch, so forgive me if I am more sarcastic than normal!
The Seven Dimensions: Get one wrong and lose $350bn! My sarcastic take on why this isn’t so easy!
1. Strategic Road Map
This is where you're supposed to hire McKinsey! Vision, alignment, a list of priority domains, clear "from/to goals," assessment of enabling capabilities, scale up, and finally, a partnership plan. This is where McKinsey makes its money.
I do have to ask: exactly where are the GenAI business cases coming from, and why do McKinsey staff know better than seasoned bankers where AI will make money? Do they channel “the force?”
Upskilling employees and attracting talent! Check! First of all, no one in the business right now can touch AI staff because of the cost, and most in play are learning as they go. I credit McKinsey with the upscaling because training an existing employee who knows something about banks will be easier than training a self-professed AI genius who has never built one!
3. Operating Model
So, this is funny. McKinsey notes that given how nascent AI is, it has been centralized in many banks. I agree; this makes sense, given the difficulty of attaining talent. Then, in a stroke of genius (why they get paid the big bucks), they recommend that as the tech advances, it should be more "federated" or pushed out into business units! Are you ready to sign the contract?
Here, McKinsey gives no practical advice other than to say that it will be "intricately challenging." What McKinsey fails to mention is that as a management consultant, picking technology is not exactly their strong point! You’re on your own!
Well, I'm feeling better knowing that McKinsey is once again emphasizing the importance of data quality, stating that it becomes even more crucial in the context of Gen AI. However, the reality is that in a survey I covered last week, 67% of risk departments reported that their data was "inconsistent or fragmented," indicating that it could not be reliably used with Gen AI.
6. Risk and Controls
Yes, GenAI has new risks like hallucinations! McKinsey's solution is to "loop in subject matter experts to validate model outputs." However, this process may not be scalable across all potential use cases with material value. You don't say. You want GenAI implemented across the entire bank and have all results hand-checked? Really?
7. Adoption and Change Management
Here, McKinsey is back in its element, where they offer the "buy 2 get 1 free" deal. Change management will be key no one doubts there will be "turf wars," as AI will not be welcomed by all. Change management is where McKinsey comes in after the Strategic Road Map!
“Among industry sectors, banking is expected to have one of the
largest opportunities: an annual potential of $200 billion to $340
billion (equivalent to 9 to 15 percent of operating profits), largely
from increased productivity.”
McKinsey’s seven dimensions are fairly obvious. What McKinsey isn’t telling you is that each one is a spin of the roulette wheel. Which tech? What talent? What do we do with our data? Each one of these can trip up even the most tech-savvy banks.
Don’t get me wrong, I think that banks will use AI effectively and make a bundle. It’s just that McKinsey’s “big numbers” reports, designed to entice consulting contracts with visions of big profits, won't work for many, even the best skilled.
So I do not think that seven steps and a consulting contract will pay off for many. My take is that, as with all new technologies the early implementations will be messy and we should not put too much faith in early leaders.
I fully believe that with AI as such a disruptive technology, leaders will come from unexpected places.
It takes two seconds to forward this newsletter and even less to subscribe!