Why AI May Spell the End of BI

Are the days of popular BI tools like Tableau, Power BI, Looker and Qlik Sense coming to an end?

If so, what’s driving this shift?

What’s different about the future of data and analytics in the age of AI?

Hi, I’m Courtney Baker and this is AI Knowhow from Knownwell, helping you reimagine your business in the AI era.

As always, I’m joined by Knownwell’s CEO, David DeWolf, Chief Product and Technology Officer, Mohan Rao, and Chief Strategy Officer, Pete Buer.

We also have a discussion with Max Votek of Customertimes about the American public’s readiness for AI adoption in healthcare and beyond.

But first, let’s kick things off with Pete Buer for a segment we’re calling In the Year 2032.

Pete Buer is back with us this week to break down some of the biggest news in the world of AI.

This time for a segment we’re calling In the Year 2032.

Hey Pete, how are you?

I’m good Courtney, how are you?

I’m good.

There was a recent post on the World Economic Forum titled, This is How Business Should Approach Reskilling for AI.

Pete, I couldn’t wait to get your take on this.

It fills right up your alley, and especially after listening to your recent discussion with David.

What were some of your big takeaways after reading this piece?

Well, thank you, Courtney.

I guess first thing I’d recommend to anyone listening is actually click through the article to the study that’s cited that it takes reference from.

Not just because all research dorks should read the study behind the article, but because there’s actually really pragmatic and useful stuff in it.

A couple of things that jumped out.

So to our title, by 2032, the article predicts 90% of jobs will be disrupted at some level by generative AI.

That is the biggest number that I’ve heard so far, and that is good because bigger is better when you’re trying to create a burning platform in an industry.

But I kind of don’t doubt it, you know, it feels right at this stage of the game.

Also useful in the study is the consultant’s favorite mechanism.

There’s a four square to be found that compares disruption to be expected within roles to the difficulty or the term that uses friction for people in those roles to find their way to new work.

So super, super useful, right?

And like within the graphic upper lefter, so high disruption but lower friction, you find roles like software and database professionals where their work is going to be disrupted, but there’s a very good chance that they can rescale themselves fairly easily into something constructive and productive.

Whereas upper lefter where disruption is high and friction is high, you find roles like office and administrative support where the roles will be disrupted and it’s going to be difficult for them to find their way to the next thing.

Strikes me as a super simple and instructive tool for leaders who have to manage transition among large swaths of employees to have a little bit of a starting point in terms of thinking about how to get from A to B.

Lots of other cool stuff in the article.

Cognizant is a staffing firm, so they have an interest in this whole notion of getting employees to their next role, helping companies fill those next roles, and they offer up a handful of their own solutions as well as industry solutions in the form of academic partnerships, company consortiums, non-profit partnerships, et cetera, et cetera.

Take away for leaders, number one, get clear on the magnitude of impact that your people are going to experience role by role by AI across time.

Get strategies in place for managing the migration now ahead of when the froth starts to hit and earn the trust of your people, and as a result, their willingness to stick it out and go through re-scaling, upscaling programs by communicating transparently what the future holds and what their options are.

So good.

Pete, I think this frame and some of the things that you talked about really helpful as you look at 2032 and all the years between now and then.

Pete, thank you as always.

Thank you, Courtney.

Is AI going to render BI obsolete?

I recently sat down with David and Mohan to talk about that exact topic.

David, Mohan, today, I wanna talk about AI versus BI, artificial intelligence versus business intelligence.

And obviously business intelligence has been around for a really long time.

And as AI continues to change the game and how business looks today, I wanna just break down how this is gonna look in the future.

Because I think it’s something that a lot of executives may not have realized, oh, this thing that I’ve counted on and used for years and years and years may change pretty dramatically.

So with that in mind, can you two break down really the evolving capabilities of AI and how that might impact business intelligence?

Before we even go there, Courtney, I would love to pull out a little bit of research.

Oracle did some research back in the end of 23 about data.

I think this is very relevant to the conversation.

Data in business, 91 percent of executives say they have so much data that it undermines their decision-making.

Yet 93 percent of executives say they don’t have the intelligence they need to make good decisions.

Like think about that dichotomy, right?

And to me, that is the essence of what you’re putting on the table.

We have so much data, we can slice it, we can dice it, we can chart it this way, we can chart it that way, we can visualize it, we can do all these things.

That’s BI.

But it just puts a bunch of information in our laps to say, hey, figure out what to do with it.

Good luck.

Whereas to me, the AI comes before the BI because it actually can tell us what to do.

It can actually infer, draw conclusions, make judgments, and bring recommendations to bear that says, hey, this is what all of this means.

This is the intelligence that comes from it.

To me, that’s what we’re talking about here today.

Yeah.

If you just go to the definitions, BI meant describing what happened in the past quarter, in the past year.

You got all of the actual data together from different systems that didn’t talk to each other, put that in a single source of truth, and then you used Power BI or Tableau or something similar to that to say, let’s figure out what happened in the last quarter.

That’s the best you could do.

I think there is going to be a version of that that’s going to be always around.

If you think of it as a financial chart, you’ll have actuals and estimated.

So BI is going to be the actuals.

You’re just truing up what happened.

AI is a whole new ball of acts that we’ve got to talk about.

It is not just describing the data, but it’s much more dynamic in nature.

You know, when you say that, Mohan, the other thing that comes to my mind is, it’s not just on top of the data and drawing these conclusions we’re talking about.

It actually can come even before it, and it is an underlying foundational technology that can be used to help understand and turn things into data that can be part of that.

So whereas BI reported on structured data mainly, now AI can actually understand the world around it.

That’s the whole thing about intelligence.

It’s learning and applying, right?

And so it actually can add value to both ends of the value chain, the inputs and the outputs.

Exactly.

Just to give a little bit of credit to BI.

In the BI world, we could do what if analysis, right?

So you could do things like, what happened if our sales went up by 10 percent or the costs went up by 25 percent?

What if this project didn’t come true or whatever it is?

You could do that kind of what if, but it seems like an audio cassette era of data, right?

So it seems so quaint what we used to do just 10 years ago.

That’s what you could do with BI.

You could do a scenario analysis.

I like that because I think if you look at the cassette player then went to the portable CD player, it wasn’t really a step improvement.

In fact, for a few of the incremental improvements you got, you actually went backwards in things like, remember how those CDs skipped and everything?

And that’s kind of like you moved from some of the traditional BI platforms to some of these modern data stack BI tools, right?

And thought spots of the world and those types of things.

Great tools, right?

But not necessarily a foundational shift in terms of how we looked at the world.

And I think you look at that continuum of how BI has matured, we’re now at a breaking point where we’re actually thinking totally different and the next generation of platforms aren’t going to be about reporting and business intelligence.

They’re going to be about actually getting to the bottom line.

What is the conclusion?

What is the decision?

And expanding the masses of inputs that we can actually comprehend, because the intelligence itself can do the comprehending in order to pull it in and make it part of the analysis.

Would y’all say it would be more accurate at this point for BI to stand for business information versus business intelligence?

You’re brilliant.

Yes.

In fact, I think it’s always been.

I think that’s totally true, right?

It’s always been about information.

It’s never been about true intelligence.

BI doesn’t learn and apply anything.

It just reports and helps us to visualize and digest.

Yeah, that’s well said, Courtney.

You know, the comparison between BI and AI quickly kind of breaks down because of some of the things that David was talking about, whereas BI was the static giver of information.

Right?

So AI is so much more dynamic in nature.

Right?

So you can think of it as a recent response to a situation that’s developing.

Right?

So you are able to kind of get to that type of reasoning.

And then you can have cognitive architectures that would give you a hand in kind of making a decision or perhaps even act on it.

Right?

So it’s a much more dynamic operating system, if you will, in its full glory.

So the comparison between BI and AI is a little bit like, you know, second graders playing baseball and major league player.

It’s totally two different things.

They may look similar in aspects, but they’re two totally different things.

So what I hear you two saying when it comes to that battle of AI versus BI, you two would not see business intelligence in it, you know, the way that we think of it today as going away necessarily.

It just becomes an input into a bigger intelligence platform.

Is that how you would say that?

Yeah, I’d say that.

I think it’s just one of the smaller blocks in the overall ecosystem.

The reason I think why we are doing this episode is from an enterprise architecture perspective.

You need to think about how information is produced, delivered, consumed.

Right?

So BI was a way of doing it.

You had quarterly board meetings and things had to get set up and BI was a valuable resource.

AI is much more dynamic in terms of operating the business itself.

So they’re all part of the enterprise architectures.

But yes, what you said is absolutely right.

You know, I would push on that a little bit further and say, I think the components of BI that remain around are going to be really merged into features of other things and bigger platforms than the BI market staying what it is today.

I think of it much like, you know, if you rewind back to the day when we had databases, and then all of a sudden, we had this application layer that came in and provided us information and workflow, and it commoditized the database, right?

And so the database was front and center in business.

It was about how we were able to access data.

You know, remember when everybody used to play with access databases and that was the tool of business?

Well, databases, whether you’re talking about access or you’re talking about Oracle or one of the modern NoSQL databases, it doesn’t matter, are below the covers, right?

And we now have this information layer of which I think BI is a major component that is what’s front and center for business.

AI is a brand new paradigm and it is going to commoditize the information layer just like the information layer commoditized the data layer.

And so intelligence will sit at the top.

And will there be components of BI that may be useful in an AI world?

Absolutely.

Right?

Some of those fundamental concepts.

But I just don’t think that market stays the same.

If I were an investor in a BI tool, I think I’d be running for the hills.

You’ve got to be able to totally reinvent because this intelligence world is going to bring about a brand new paradigm.

And it’s only those technical components that will be leveraged by AI that I think will remain.

So what I hear you saying is AI is the end of BI?

I think for all practical purposes, yes.

Right.

So it’s not, as David said, the elements of it is going to remain in the enterprise, but it’s for all practical purposes, it’s a new paradigm for one reason.

And if I have to give is around AI has a memory state.

Right?

Your future action can be predicted based on what happened just before.

That is, BI is nothing like that.

So two different paradigms.

BI will remain in some corner of the enterprise, but it’s two totally different things.

I love how you said that, Mohan.

I think that’s totally right.

The memory state matters a lot in this.

I think the input, not just the output, is the other piece that’s going to change the game here.

If you think about how much it currently costs to do a BI implementation and all of the data warehousing and the data lakes and the data engineering work that has to be done, we hear all the time how painful that is for organizations, how costly it is for organizations.

We’re talking years of time and millions of dollars and still not getting it right.

The juice is just not worth the squeeze.

Whereas with AI, I think we are rapidly getting at a point where we can understand and make sense of data much more easily without all of that cost.

Yeah.

Well, David, Mohan, thank you for helping us think about this topic and how the business platforms we’re so used to may be changing here in the very near future.

So as I wrap this up, you know, the two big things that I hear you say is the next generation of platforms won’t be about reporting.

They’ll be about providing proactive intelligence and helping you make sense of the data or the information.

And the second thing is AI will provide more dynamic real-time operating system that will transform what we currently know as the information layer.

Thank you, as always.

Thanks.

Hey y’all, Courtney here.

Because you’re all loyal AI Knowhow listeners, we want to reward five of you with a special limited time offer of three months of free access to Knownwell.

Now, because there’s a significant investment on Knownwell’s end to onboard new customers, we do have some qualifications we have to meet.

So if you’d like to see if you qualify, drop us a line at Knownwell, at knownwell.com, with the subject line AI Knowhow Offer.

And someone with our team will get back to you as soon as possible.

That’s knownwell at knownwell.com.

Max Votek is a managing partner at Customertimes, a global engineering product development and technology consulting company.

He sat down with Pete Buer recently to talk about a Customertimes study that found the general public is much more open to AI than you might think.

Max, so great to have you on the show.

Welcome.

Hey.

Hey.

If you wouldn’t mind, could you start us with an introduction to Customertimes and your role there?

Yeah, sure.

Customertimes is a global company specializing in digital engineering, product development, and technology consulting.

Headquartered in New York, so our team comprises 1,500 experts worldwide, providing tailored technology solutions for substantial business returns.

I transitioned from being a pharmacist to entrepreneur and co-founded Customertimes.

Currently serve as a managing partner, leveraging my extensive expertise to guide the company’s strategic direction and ensure the delivery of high-value solutions for our clients.

I understand because our team tends to stock the attendees to the podcast.

I understand that you’ve recently done some work at Customertimes on researching public perception of AI in healthcare.

And we’re hoping that you might be able to share some of the highlights of that research.

Basically, in our study at Customertimes, we surveyed 2,000 Americans to uncover the real public’s perception of AI in healthcare.

So results indicate a really promising outlook.

Forty-eight percent of respondents are optimistic about AI’s role, with 43 percent believing AI will enhance the diagnostic accuracy.

So moreover, a quarter think AI will reduce the healthcare costs, and 56 percent feel AI design drugs will be cheaper.

Significantly, two-thirds say AI will outperform humans in diagnosis and treatment with 40 percent to follow AI-generated medical advice.

Nearly half think that implementing AI in healthcare will help cure deadly diseases like cancer.

And interestingly, the 40 percent believe AI could replace doctors, though only 10 percent think it should, indicating a nuanced perspective on the future of integration of AI in healthcare.

All right.

There’s a lot going on in there.

I’d like to try to pick some of it apart if I could.

First, let’s start with that last stat, that there’s a substantial percentage of the population that believes AI could replace doctors.

But even more surprising to me, that there’s some not-insignificant number who think it should.

What is the logic?

What’s going on behind that?

I mean, artificial intelligence, having roots in early developments, is gaining popularity only now.

Like any new technology, it faces a variety of perceptions, both positive and negative, which is entirely normal.

So historically, the Industrial Revolution faced similar scatolicism.

Many feared machines will eliminate jobs, but ultimately increase efficiency and productivity.

So if you take any aspect in the healthcare, some of the processes and some of the thought process of the doctor can be transferred to parallel processing in AI and make it faster for the doctor, obviously.

But at this point of development and at this point, at any regulated environment, there’s very little potential to eliminate any of the doctor tests.

But you can eliminate the burden on the doctors and eliminate the feeling of the burnout of the doctors in the daily routine.

Health care is a complex service.

Lots of costs associated with it, lots of specialization, lots of humanity in interactions.

I might have said that would have been one of the trickier industries to see positive uptake of AI.

Do we think that says hopeful things for AI in other areas of business?

Yeah, I think for doctors and for health care, so AI would bring numerous benefits for both doctors and patients.

So for doctors, it’s always about improved diagnosis, accuracy, so an algorithm can analyze medical data and imagine more quickly and accurately than humans leading to more precise diagnosis, always checked by a human being, of course.

So it might also help a lot with the time efficiency.

AI can handle the routine tasks such as data entry, appointment scheduling, patient follow-ups, allowing doctors to focus more on patient care.

And it’s also about enhanced decision-making.

So AI can provide evidence-based recommendations and predict patient outcomes and assist doctors to making better decisions.

We also need to add that predictive analytics, what we can see is, can predict patient deterioration, disease outbreaks and other health trends, enabling preventive measures and timely interventions so the doctor can make better decision faster.

So, as well as we see within our customer ecosystem, that now AI is optimizing hospital operations a lot, so managing inventories, eliminating inaccuracies in invoicing and allocating the resources more efficiently.

How about in the data, were there any places in the chain of activity in health care that you tested, where public perception was less strong, like places where people might prefer AI not play?

I think the public perception has hesitations about almost everything.

So if you ask this question, like giving the opportunity for an AI to make the decision instead of the doctor, so probably send some hesitation to a human being.

Although, giving inspiration for the doctor to think about the potential diagnosis and other, even reading through the lab results, so AI has much better potential than a human being to translate this into action and to give the advice to the doctor to think about certain things.

While looking at the lab results, so the pattern recognition, so AI algorithm can identify patterns and correlations in lab results that may not immediately be obvious to human doctors and helping to detect abnormalities, or trends, or indicative, or specific conditions, right?

So it also can compare the current lab results with historical data and identify the significant changes or trends over time.

So typically, if you go to the doctor, if you visit the doctor, so the doctor has some of your medical history, but not all of it always, right?

So imagine it having the entire history from years and years and years and years.

So typically, the doctor visit would last like, let’s say 15 minutes, 20 minutes, 30 minutes at the maximum.

And imagine the doctor being able to look through all your history, like 10 years, 20 years from now.

And it will definitely increase the quality of the decision-making if there are some abnormalities in the past, which can help to make a better decision now.

So that’s what we see.

That’s the potential which can be easily transferred from predictive analytics models from other industries, like Telecom, or like Churn Prediction has the same model that can be applied for the health care, like the same mathematical models can be applied and can increase the quality of personalized care and increase the quality of each visit that you go to the doctor and you get much better care.

Sorry, AI as an input to the system as a co-pilot, the world’s not quite ready, however, for AI as an independent agent.

Awesome, and thanks for sharing all the different use cases, the places where AI can be leveraging data and improving processes.

Heck, if the doctor just remembers what we talked about last time we met, I’ll consider it an improvement.

Well, we are grateful to you for participating and sharing your insight with this particular source of information on AI.

Thank you for being here today and for spending some time with us.

Thank you, Pete.

Thanks as always for listening and watching.

Don’t forget to give us a five-star rating on your podcast player of choice.

Hey, go to knownwell.com/demo if you’d like to get a guided tour of our AI-powered platform for commercial intelligence.

At the end of every episode, we like to ask one of our AI friends to weigh in on the topic at hand.

So, hey, ChatGBT, what’s happening?

This episode, we’re talking about why AI might spell the end of BI.

What do you think?

I think AI could really disrupt BI by automating data analysis and generating insights without human input.

It might make some traditional BI roles obsolete as AI takes over data crunching tasks.

Now, you’re in the know.

Thanks as always for listening.

We’ll see you next week with more AI applications, discussions, and experts.