WATCH: Cognitive Automation Is The Brain of Your Supply Chain

By
From The Editor
60m
WATCH: Cognitive Automation Is The Brain of Your Supply Chain

Aera CEO Frederic Laluyaux shares his thoughts on cognitive automation and its powerful ability to scale decision-making

John Kao:

Hello, Brian.

Brian Solis:

Hey, John. What are you doing here?

John Kao:

I'm improvising. Madly improvising today. It's been one of those days.

Brian Solis:

It has been one of those days and one of those years where we're all figuring it out as we go.

John Kao:

One of those lives.

Brian Solis:

It's good to see you, John. Last week I think I said this and it is truer this week is it was yesterday that we just did this show. I mean, it was not,-

John Kao:

Not possible that a week has gone by.

Brian Solis:

Oh, my goodness. Well, with that said, let's jump right in because our guest is backstage and they have a hard stop and I know that I could banter with you all day. Well, welcome, everybody. Welcome to Intersections. Let me just go ahead and kick things off so we can jump in. This is a community whereas John and I and the guests that we bring together, plus your involvement, your participation, your comments, your questions, the things that you share in between episodes are all what make Intersections such a special community.

Brian Solis:

Welcome. Intersections, for those who are new, it's a weekly conversational jam session, jam session that dives deep into the intersections among technology, innovation, culture, and ideas. We bring diverse personalities and world views together in the service of greater understanding and unlearning. And look, this is as much about innovation and transformation as it is about artistry, creativity, empathy, and all of the things that we need to celebrate now in a world without a playbook.

And with that said, I'd love to introduce you to my co host and co-founder John Kao, whom the economist is called Mr. Creativity and a serial innovator. A man whose pedigree, whose experiences, whose work, and whose network are just unbelievable. And we see that each and every week with the guests and today is no exception that he invites onto the show some of which are former students.

He's a thought leader, entrepreneur, advisor, who's played a prominent role in the fields of innovation and business creativity for over 30 years. That hardly does his portfolio justification, but I'll do my best. He's taught at Harvard Business School. He served as visiting faculty for MIT, media lab in Stanford.

He's attorney, nominated producer film and stage, and he wrote the bestselling book, Jamming the Art and Discipline of Business Creativity. Mr. Kao also writes for Forbes. Please do check out his column. It's riveting actually and do follow him on Twitter and LinkedIn. John, welcome to Intersections.

John Kao:

Hey, Brian. I'm all warmed up now. That was a great intro. Let me quickly reciprocate by saying that Brian is my colleague and partner in crime on matters innovation because innovation we know is in a constant state of need for reinvention and for applying 21st disciplines to a phenomenon that was first really described in business going back to the 19th century.

So Brian is a polymath communicator, domain expert, immersed in things digital before digital was really noticeable, agenda item on people's radar screens. So digital and social media, digital and marketing, digital and innovation, digital and organizational change, digital and leadership. He publishes, he speaks, he shares his knowledge, he advises and in his spare time he has a full-time day job, which is evangelist for innovation at Salesforce. So, Brian, here we are.

Brian Solis:

Here we are. And here we all are together. So please do share where you're tuning in from, where you are in the world. We'd love to see our community global come to life every single week that we get together. And do also share your comments and observations and your questions. We'll do our best to feature them on screen and to get everything commented on and answered as we go. John, because we are on a time schedule I'm going to throw it over to you to introduce our first guest.

John Kao:

So Stephen Ezell is someone that I've had the pleasure of working with off and on for almost a decade I would say. Stephen is the vice president for innovation policy at the IT Innovation Foundation based in Washington, D.C., which is why he is wearing that nice blazer as opposed to this unconstructed archaic hiking jacket. He is the co-author of several books.

One with the head of ITIFs programs Rob Atkinson on innovation economics. He has a storied past as a serial entrepreneur and someone who's delved into financial services degree from Georgetown with regard to foreign service, which we all agree is more important than ever at an era where the room for improvement seems infinitely large and get larger by the minute. So, Stephen, welcome. Thank you for being here and thank you for being here on short notice.

I spoke to Stephen last night probably around 8:00 or 9:00 his time and he graciously agreed to attend. And part of this is because he and I share a passion for innovation and science and technology policy in the U.S. We actually collaborated on some of the work that is somewhere in the congressional process around what the next wave of funding and programmatic activity is.

And I thought it would be a great opportunity to get Stephen who is up to date on these things like nobody's business to really engage with us in a conversation about what the heck is going on. So, Stephen, with that wordy introduction, welcome.

Stephen Ezell:

Well, thank you, John and Brian. It's a pleasure to be back with you all.

Brian Solis:

Welcome, Stephen.

John Kao:

So I guess I'll start off by saying, what the heck is going on? The last I recall there was a big piece of legislation, gazillions of dollars, 10 innovation hubs, a sharp negotiation around making sure the word innovation got included in the agenda and hope spring at the tunnel with regard to an American National Innovation agenda. What's the latest? Where are we? And what's been progress and what's been non progress?

Stephen Ezell:

Well, John, what you're referring to is the United States Innovation and Competitiveness Act, USICA, which was passed early in the summer by the Senate on a bipartisan basis, 68 to 32 for a $250 billion legislative package. That among other things included $120 billion of new proposed R&D investment over the next four years, $50 billion of act to restore Americans semiconductor manufacturing, other titles in there to fund DARPA, et cetera.

A program called a Global Economic Security Strategy to have a comprehensive approach to China. So really positive bipartisan action from the Senate. So where does it stand now? Well, of course, action has to be taken up in the House and what's happened in the House is a matter of bureaucratic jurisdiction. So one single bill in the Senate now has as many as five to 12 different House items that are coming out of health science, house commerce dealing with various facet of the bill.

You mentioned the regional tech hubs, there's regional Tech Hubs Act. There's a reauthorizing NIST Act. There's funding for NSF Act. So essentially to answer your question, what is ultimately going to happen is that these various pieces of legislation are going to come out of the House and then a conference is going to be established between the Senate and the House probably later in the fall, sometime before Thanksgiving.

And they'll sit down and they'll try and pull these packages together. I can also tell you talking to folks on the hill that right now infrastructure and the budget reconciliation have priority. You'll see action on those things before you see Congress coming back to this legislation later in the fall.

John Kao:

So let me understand a little bit more the details of the process. So you said that in the House the unitary legislation that was approved by the Senate is now deconstructed into agendas that relate to different interest groups in the house and that there will be some kind of a conference to essentially prioritize, decide what's in the tent, what's not in the tent and what should be funded. Is that more or less accurate?

Stephen Ezell:

That's exactly right. So whenever you have differing sets of legislation, it can be the same topic, but the House has got to bill, the senate has got a bill. When those entities are passed by the respective bodies then these conference committees come together and they hash out a final version that would be sent on to the president for signing.

John Kao:

And I'm fascinated by this word conference. So, I mean, what does that look like? If we were looking through a one way mirror, I mean, is it a whole bunch of people from legislators from the house and Senate sitting around some mahogany table or some virtual mahogany table with a whole bunch of back benchers and debating, I'll give you this synthetic biology piece for Iowa if you give me... I mean, is it horse trading or what is it likely to look like?

Stephen Ezell:

Absolutely. If you think about how legislation within a body gets created, what are called markups. This is when a House committee will meet to discuss a piece of proposed legislation and it's marked up with changes. Essentially what a conference is, is a markup action between the House and the Senate at the same time. In terms of the horse trading, I think an awful lot is going to be directed by what the White House indicates to its congressional allies.

It views as some of the most important and legislative priorities. Also from what I understand, majority leader Schumer has really made USICA the key part of his legislative agenda legacy for this conference, preferred this Congress, right?

So I do think, and also the Senate being in the upper body, that many of the priorities that the Senate will have, especially around things like the Chips Act, the regional technology hubs, that will ultimately see those come forward in some capacity.

Also think about it. We're talking about a $3.5 trillion infrastructure package and much needed things, improvements for roads, sports, education, college, community colleges. I think if you're going to see that invested then we can do at least $120 billion for R&D.

John Kao:

Right. It would seem only fair, right?

Stephen Ezell:

It would seem so.

John Kao:

So how much visibility is the public going to have on these deliberations in the conference and also in the run up to the conference where I'm understanding now that there will be this number of deconstructed modules or agenda items that need to be discussed. I mean, are we going to get to know what those are?

Are we going to have some visibility over the horse trading, or is this going to be essentially kind of how much, I don't mean transparency in the sense that information is being concealed, but more just understanding what the process is in terms of what visibility the public is going to have over it.

Stephen Ezell:

Well, one thing I think you'll see the House pick up on this fall that, in fact, the Senate already did in the spring, whenever you have this type legislation there's a number of hearings, right? So I suspect I just have before me here are five pieces of legislation that the House is now considering. The Regional Innovation Act of 2021, the Energizing Technology Transfer Act of 2021, the National Science and Technology Strategy Act of 2021. All these are the various facets of things that are already in USICA.

What I'm saying is that on each of these pieces of legislation you'll see the relevant House committee have hearings in the fall. And, of course, those public hearings are an opportunity that the public can submit comments to congressional staffers that you can hear from experts on the issue. And that's the opportunity the public will have to weigh into the policy making process.

John Kao:

It occurs to me it would be really useful if you and I could collaborate on a short piece, which I could put in my Forbes column about sort of a viewer's guide to how this innovation dialogue, or horse trading, or conference, whatever term we want to use is going to go down.

So that rather than learn all about it at the end of the year when it's all done we get to look at some of these agendas, know what the public hearings are and things of this kind. Because I know for me I got privileged access to you. So I get to ask you what's going on, but most people won't be able to figure it out.

Stephen Ezell:

I'd love to collaborate there, John. I think it's also important, if I may-

John Kao:

Yeah. Please.

Stephen Ezell:

... to talk about why this type of legislation is important. Because you talk to many and they'll say, well, doesn't the U.S. just lead the world in innovation? Isn't leadership and innovation our birthright? Aren't we number one in everything? You look at the Forbes, or the Fast Company or the Boston Consulting Group, list of the world's most innovated companies and two thirds of the top 50 are always from the United States and seven of the 10 from that was in California, right? So at the firm level we do very good.

But if you look at national rankings, indices of innovation, like the Bloomberg Innovation Index, whereas the United States was number one in the world in 2013, we've slipped to 11th in their latest 2021 ranking. So the reality is that while the U.S. has many innovation strengths, we also have some quite significant innovation weaknesses. For instance, if you look at how much our federal governments invest in R&D as a share of GDP, we actually invest less today than we did before Sputnik.

So as a share of GDP, our federal government invests less than we did in 1955. We're seventh in the OECD, a group of nations on our national R&D intensity. So we're not investing enough relative to our size compared to our peers as well as to our past norms. And this means that we're not providing the seed corn that future generations will need to create the new technologies, occupations, industries of the future.

And I think another very serious concern is when we look at inclusiveness of opportunity in the innovation economy, in 2019 my organization did a study with Brookings where we looked at where innovation jobs are located in the U.S. economy.

And we defined those as jobs and industries that essentially have a certain share of STEM, science, technology workers in their workforce and the level of R&D they invest. Essentially what we found is that one third of innovation jobs in the U.S. economy are concentrated in just 14 counties.

John Kao:

Repeat what you said.

Stephen Ezell:
One third of U.S. innovation jobs are found in only 14 U.S. counties.

John Kao:

Counties.

Stephen Ezell:

We are San Francisco, San Diego. Half of jobs in just 41 counties. And that the disparity has massively increased over the past decade. And that is why parts of this legislation like this regional tech hubs are designed to provide money to other potential up and coming innovation hubs in the U.S. economy, places like Manchester, Nashville to empower their workforce education, their university technology transfer systems and so we can have more innovation nodes across the economy.

John Kao:

How much is the location? Well, first of all, I mean, I'm hoping that the regional innovation hubs portion survives, but then the question is what's the process by which the negotiation is going to happen around which 10 places get it or X number of places gets anointed?

I imagine that there's a checklist of university present, yes. Indigenous tradition of some kind of technological innovation, check, business-friendly climate, check, et cetera, but still that narrows it down to maybe 50 places. So is the conference part of the horse trading for those locations or is there going to be a truly rational process for doing it that takes into account national, not just parochial local agendas?

Stephen Ezell:

Right. So USICA provide $8 billion from 2022 to 2026 for such a program. Now that's just funding. The program itself would be administered by the department of commerce and DOC then would set up a competitive grant evaluation program. There's actually been some other stuff we've done under the Obama administration.

We created something called the Investing in Manufacturing Community Partnerships, the IMCP program and this designated 25 areas for federal grants to stimulate the manufacturing economy, do swat assessments. So some of these structures that commerce has had in place for evaluation of these applications before will probably resurface.

John Kao:

So without putting you on the spot, I mean, how do you feel about where we are or where we're likely to wind up at the end of the year or early next year in terms of our government taking a more active role in terms of lighting the boilers of innovation science technology et cetera.

I mean, are we excited about it? Do we feel we've got a half a loaf is better than none? Where do you think we are relative to where we need to be? Because I think nobody argues with the idea that the room for improvement is infinitely large and what's your assessment of where we are?

Stephen Ezell:

I think should all these programs come to pass, the infrastructure bill, the USICA, then we will be making the types of investments we will need to be able to compete in the future. I think then what the challenge will come is around implementation of these programs across the various federal agencies. And can we, in fact, do that in such a way that within two to four years is able to document success and impact?

Because a concern I do have is that if we do throw a lot of money at some of these challenges, but it's not effectively implemented, then we're going to find ourselves in situations several years where the naysayers about industrial policy or federal support for innovation will say, wow, look at this.

Where is return on this type of money? So I think that at least this point politically, obviously the Biden administration has both houses of Congress. They got energy behind this. This is all a part of President Biden's broader build back better strategy. So I think it'll happen. We'll see if it delivers the success that people hope for.

John Kao:

One last question from me and then I'm going to throw it over to Brian. How much does the narrative around our competitive posture with China and the China threat, if you will, play into the momentum behind all of this innovation?

Brian Solis:

Great question, John. That's a great question.

Stephen Ezell:

Well, I don't think there's any question that the increasing recognition that China poses as a competitive threat to the U.S. economy has been a central animating force for the bipartisan support line and legislation.

You would not have individuals, in my opinion, such as Rubio, Cornin, Graham aggressively supporting some of the arguably industrial policy oriented elements of these technology strategies if not for an increasing realization that we need to enhance our competitive capacity as a nation to meet the threats of the future.

You look at China's Made in China, the MIC, 2025 strategy calls for $1.7 trillion of investment in 10 critical technology areas. Artificial intelligence, supercomputing, genomics, right? All these things that we're talking about. That's a shot across about and I think more people in Washington realize that.

John Kao:

So that's an order of magnitude bigger than what we have on the table basically.

Stephen Ezell:

Yes, exactly.

John Kao:

That's sobering fact.

Stephen Ezell:

It's 10 times more.

John Kao:

Yeah. Order of magnitude. That is sobering. So on that happy note, I'm going to invite Brian into the conversation because I can see the steam coming out of your ears.

Brian Solis:

I thought it was a filter. Stephen, can you share that stat one more time please? China's 2025.

Stephen Ezell:

So China a couple years back created that's kind of Made in China 2025 strategy, which clause out investments in 10 strategic areas of their economy. But there are in total calling for now $1.7 trillion of investment in those 10 technology areas.

Brian Solis:

And that's 10X over the United States. Is that right?

Stephen Ezell:

Yes, exactly.

Brian Solis:

I asked for you to reshare that because I wanted number one, to make note, but also two, for those watching here to demonstrate... There's a scene, I can't remember what it was from, but it talks about how there's a student, she gets up and she talks about how the U.S. is the greatest country in the world. And then someone on stage says, actually we're not the greatest country in the world. We're number such and such in math, we such and such an innovation.

And that too was sobering, that stat of a reminder that we aren't the greatest country in the world. And we're actually in a serious position at risk of losing not only that position, but actually the sentiment and the ideology around that position. So what I'd like to just quickly share, and then I'll jump to my question because I know you have to go, I presented in China around digital transformation and innovation in what's largely considered Shenzhen, an innovation hub for the country, like Silicon Valley.

And the nature of work that I had to do to tell a story of innovation to a country investing in innovation at that level was far different. And what I want to share with people, and then I'll jump into the question, is that China actually sees innovation as table stakes. And then the rest of this stuff is how they make it happen. In the United States Innovation is a cost center.

And when we look at how we invest in it as a country, there are people who get it and then there are people who politic it. And the infrastructure bill is just one example of that. So, Stephen, I'd like just to close this out with your advice for how every day leaders in innovation, and so those are entrepreneurs, those are investors, those are people with the stake in the future of technology in the United States how they can become more involved with assisting in the furtherance and the acceleration of these policies.

Stephen Ezell:

A couple years back, if you recall during the Biden administration when we had Republican Congress, we had this thing called sequestration was across the board cuts to federal expenditures so we could balance the budget. And while that has merit as an enterprise one thing we were suspending was kind of investments in R&D. So I was actually on Capitol Hill and I was talking to a U.S. Congress member and I was trying to make the case that, hey, we can't cut the R&D funding. We have to continue to invest in the future.

And this gentleman said to me, but Steve, isn't innovation the problem? Isn't innovation why we have these technologies that are costing workers their jobs, isn't financial innovation [inaudible 00:25:08]. So I think, to your point you made about China, restoring the belief of our society in the potential and the promise of technology and innovation to deliver a better future is a key challenge. Now to do that, we're going to have to make it broad and inclusive.

That means we're going to have to educate our society with the skills they need to be competitive in a fiercely competitive global economy, right? So we got the education part of the equation. We have to get our firms to understand that there are needs to invest in the ecosystem. in 1984, the national association of board of corporate directors changed the fiducial responsibility of boards of directors from what happened before workers, the community, shareholders, employees to just one thing, shareholders.

And I think that has led to a degree of corporate share in the United States. So across the board we need to reanimate our societies' understanding of the need to collaboratively invest in our innovation potential. And I think that would go a long way. And I encourage your viewers here in the U.S. to talk to their policy makers, talk to their Congress members, have them emphasize the priority of investments in infrastructure, in R&D, in education and the things we need to have a platform to compete in the future.

For your viewers in other countries I would encourage them to do the same thing and also to encourage policy makers to only focus on good innovation policies. So some countries in the world try to win an innovation through IP theft, or force tech transfer, or excessive subsidies. We need to encourage global policy makers to only put in place good policies that ratcheted up the quality of the game of global innovation because they're predicated on fielding stronger teams.

Brian Solis:

Well, there right there is the playbook for moving forward. Steve, I know you got to go. So we'll just say goodbye and thank you so much for joining us. Always a pleasure.

Stephen Ezell:

I always enjoy it. I hope we have the opportunity to do this again. And I look forward to the next episode of Intersections. [crosstalk 00:27:19].

Brian Solis:

Thank you. If you talk to the Biden administration let him know that John and I are here to help.

Stephen Ezell:

Sure thing. Will do. Thank you guys so much. It's been a pleasure.

Brian Solis:

Cheers. As always, John, he's incredible. Just kind of makes my blood boil. Here we are arguing and debating about some of, the only word I could use, some of the stupidest things day-to-day where in fact there are some major things that are really going to impact this country and its position in this world beyond the world itself.

But with that said, John, I want to hear you riff a little bit and react to his last statement on how we can all get involved. And I'm going to multitask and make sure our next guest knows how to get in.

John Kao:

Well, if you hear drilling in the background this is the test of how directionally capable this show microphone is. Look, I think that innovation won't really become a relevant national agenda. It's just something for elites and people who are specialists and legislators, academics, entrepreneurs, venture capitalists, et cetera. Innovation really needs to be embraced by the rank and file of our entire country as a national agenda.

Much as in the past in our country, there have been periods where people got behind the space program. When we landed a man on the moon, they got behind. The industrialization of America during Second World War, they got behind, the notion of science and technology when Sputnik went up.

And so one of the things, Stephen and I co-led a process called reimagining American innovation back last year where we brought in people like Senator Schumer, chief economic policy advisor, and a bunch of domain experts in this whole area of large scale innovation.

And one of the key priorities that we decided converged on was the notion of a national narrative, that there needs to be a "why" for innovation as well as a "what" that resonates with everyone not just with, as I said, people who have innovation on their business card.

Brian Solis:

Yeah. Absolutely. In fact, thinking about it too, that I know our next guest is backstage, so I'll be quick is that there's a narrative, there's a story that needs to shift and change to invite those who aren't, let's say, what's the nicest way to put this, say illiterate, I guess, in the meaning of innovation and its benefits to society and people and not just economies.

And as David called out, the David Gonzales here in the comments, from an outside in view, it looks like the United States is more concerned in short-term revenue and profit. And I think that short-termism is another pandemic that we do suffer from it. It's economics. It's the model. It's the capitalism. And it's also what's driving a lot of our policies right now and a lot of our division.

John Kao:

Well, I want to say something about that, because I think that it has to do with a sense of time and a sense of history. We are a culture that values the ability to turn on a dime, jump on an opportunity, make something happen, be spontaneous, improvise, et cetera. And as a result, we sometimes get, as the British would say, our knickers in a twist, or we ignore the long-term implications of our actions.

And if you think about China's culture, for instance, China is used to thinking in millennia and dynasties. And they're not thinking about the impact of made in China on the next five years. They're thinking about it in terms of the impact in the next 100 years, the next 200 years and what the world is going to look like. And if we don't start getting some longer term perspective into our strategic planning as a country, we're going to pay for it.

Brian Solis:

Absolutely. And we're starting to see those costs come in due, and receipts will follow. And with that said, actually, John, maybe we talk about this individually on another episode because you just got me thinking, for a nation that's viewed as agile, you and I both spend a lot of our time working with executives and policymakers who are anything but. So rooted in the past and the way that they know the world.

And in fact, it's sort of the antithesis to that impression. But with that said, I do want to bring on my friend and someone who is changing the world and has changed the world in the past. So he also is a serial entrepreneur and a serial innovator. His name is Fred Laluyaux and he is the president and CEO of a company called Aera Technology.

And Aera is a company that I had the opportunity to work with. I've had the opportunity to work with Fred. That is creating the cognitive operating system. Essentially creating a cognitive enterprise that will eventually become self-driving. Now with that said, I'm going to multitask and bring Fred on. Fred, welcome to Intersections. How are you, my friend?

Frederic Laluyaux:

I'm good. How are you?

Brian Solis:

I'm good. I miss you.

Frederic Laluyaux:

I know. I miss you too. It's been too long.

Brian Solis:

It has been too long. Fred, we'll just jump right in. Thank you for taking time out. Why don't we start from the basics? For those who are watching the show, I actually invited a bunch of colleagues from Salesforce onto this this particular segment, because they're working with a lot of CPG companies, a lot of retailers who are trying to figure out how to navigate future disruption in the supply chain. It's a big deal. It's not one that really makes the 8:00 news, so to speak. So Fred, tell us about Aera and what you're working on, and we'll go from there.

Frederic Laluyaux:

All right. So first of all, thanks for having me today. And John, good to meet you. We haven't had a chance to speak before. So Aera is a project that has been in our heads with Shariq, co-founder, and myself for quite some time. And Brian, you got exposed to it quite early on. The high-level idea is the realization over the last 10, 15 years that the demand for accurate, timely, intelligent decisions is increasing the supply.

It's very simple supply demand problem. We're trying to provide tools and data and systems for people in large organizations to make better, faster decisions closer to the point of impact. And every time we move forward, the demand moves forwards faster. So there is a gap between supply and demand. It's not supply chain, but it's literally supply chain of decision. And a way to fix that is to bring a digital brain that can actually bring the agility and the scale in the decision-making process.

And that's what Aera is all about, is they provide building and delivering this digital brain that will allow organizations to decide fast and accurately enough to catch up with the disruptions in supply chain, but that disruption in supply chain, we just look at it as short term COVID thing, but it's true, it's there. But it's really driven by some profound changes in the way we buy, in the way we consume, in the way the world works.

And I'm sure you've talked about it extensively in this show. So Aera is born a few years back with the idea of providing that digital brain that would bring those organizations the agility and the scale in their decision-making process. Interestingly enough, started in the world of supply chain, but today, being pulled in revenue management, in sales, in finance, because the problem of supply and demand in decisions is everywhere.

Brian Solis:

Yeah. That's fantastic to hear. I always had a sneaking suspicion and a gut feeling that Aera would become the brain of an organization at that level beyond supply chain. But the interesting thing about it is that you and I were working together before COVID and then a lot of things changed in March, 2020. So I want to back people into Aera and its value proposition.

So if you don't mind, I'm going to just kind of take you back a year. And let's talk about what is that brain in the organization, and let's apply it to supply chain and let's apply this example in, for example, how a CPG brand might know how many things to create sell price and deliver with an analog brain, and then the role of a digital brain in a cognitive enterprise as a result of getting better.

Frederic Laluyaux:

Yeah. So if you look at the decisions that companies are making, there are some strategic decisions, there are some long-term planning decisions that require a lot of data, and knowledge, and tribal knowledge, and math that is applied to that decision. And there is also a third category, which is the decision that you make on a daily basis. Should I ship this by freight, by road? So-and-so forth.

Should I hold this order? Or should I push this order? All of those day operational decisions that you make on a daily basis are made by people. And every decision looks the same. There is an event. I need to analyze some data, pick the event and then deploy some logic on top of it, right? So let's take a very simple order example that everybody can understand around I look at an open order.

Well, I need to match it with inventory. So humans will go and say open order, matching and inventory. Do I find this in inventory? Yes, no. The answer is no. What do I do? Well, do I find the inventory somewhere else? If I don't find it, then can I predict excess inventory? And you start going through a logical series of steps are always relied on data, on science, basically a set of operations, optimization, allocations, predictions, and so on and so forth.

A flow that drives the logic across data and people, and some validation steps. And then there is an execution step. Once the decision is available, you need to execute it, right? So this is a very tedious process. This is what our dear friend and is on our board, Joe Fuller from Harvard Business School called a lot of the dole work, right? It's repetitive, and this dole work can be actually automated.

Now, there's value when you do that. You free the operators from doing that dole work, you expand your range and your scale. Every organization has classifications. I like to take that as an example to illustrate the problem, right? So if you work for a big CPG company, you were talking about CPG, they have ABCD classification, XYZ. We classified the product, the customers, the market opportunities, why?

To basically make sure that you deploy your limited decision making power to the products, to the customers that matter the most. That's why you do classification. But imagine that you bring this scale, this digital scale, this brain that can repeat 24/7 those operations. The need for classification goes away. So we get immediate scale in your decision making power. That's what we call cognitive automation. But there's another part, which is the augmentation.

By doing so, our technology is able to create a permanent memory of all the decisions that are made on a given problem, either the decision is made by the system itself or with the interactions with the operators. And that permanent memory can then be leveraged to improve the quality of the recommendations, the quality of the decisions. And it's a complex problem because is it a good decision or a bad decision?

Well, it depends how you are looking at it. Are you looking at it from a cash perspective, a cost perspective, a service level perspective, a sustainability perspective, a risk perspective, a compliance perspective? So there is not good or bad. And this is why you need data science to really help make sense of that data set. So we're talking about automation bringing the scale and the agility, but also augmentation, which is bringing better decision, again, defining on how better.

And the other thing I would say is the problem right now is getting worse for large organizations, despite massive investment in all sorts of technology, data lakes, and planning tools, and business intelligence, and AI because the complexity of the decision is increasing exponentially, not just the volume and the speed at which they have to be made. All the cycles are accelerating.

And we saw that during COVID. But also the complexity itself. If you're doing a procurement decision, used to be quite simple, one or two dimensions. Today, you're looking at multiple dimensions. Am I buying from a vendor that is sustainable? And so on and so forth. And this, if you want to do that in real time, requires connecting countless data feeds and logical flows into a single digital process.

Brian Solis:

Yeah. That was so eloquent, Fred. And I can now see why the acceleration beyond just supply chain into all other facets of the enterprise have become both accelerated and inevitable, and that's fantastic. Because I remember the conversations you and I used to have around the analog brain and bringing 30 years of expertise in order to be able to make these decisions. Number one, it wasn't fool-proof.

Two, there was no systemic memory or knowledge within the organization to build upon that. So now with the digital brain, you create essentially a cognitive operating system that can transform function by function, but ultimately over the enterprise, to become self-driving. And before I turn it over to my cohost and friend, John, can you help us connect the dots between where we are today and the idea of your vision for what a self-driving enterprise becomes.

Frederic Laluyaux:

Yeah. That's a good question. Well, it's real. It's very short answer in three words. It is actually real now. We've deployed that technology. Now, we didn't do anything. Our customers have deployed our technology in some of the world's largest companies. And it's real in the sense that we now have systems running in full autonomy. Literally the system detects the events, deploys the logic, and takes action. So the system is able to ride back in the transactional system. So that's real.

We've done it at scale. Billions of records are being pulled into the system and the digital brain every day. We're still learning, which is so much fun. I mean, it's rare in our industry that you're building something that's not been done before. It was usually, in our software industry, we kind of doing a incremental improvement on something that already existed. Here, it's never been done.

So how do I build a trust between the digital brain and a business operator? Because today, Brian, if I come to you and say, I recommend that you ship this this way, you're going to say, Fred, I trust you. I know we've worked together. And I know that you're looking at the right data, and I know that your logic is good or bad, and you will have a debate. Now, imagine that it's not me coming, but it's on your phone, like a Siri or Alexa.

You're getting an interaction with a system that says, I recommend you do that right now. I need to establish that trust and basically digitize the next level of double click. What is the data that you're using Aera? What is the logic that you're using? And establishing that trust across a variety of skills. So we had to build a system of engagement to really connect the users. And the analogy here is Siri. With Siri, I engage across multiple skills, turning on the lights, putting some music, giving me directions.

And it's the same system of engagement, right? You don't have one Siri for every one of these skills. So we had to create the same thing to build an environment that allow the trust between the business operator who might be exposed to a decision that has to be made based on his knowledge and a digital brain. The second thing we had to do, and that's also very real, is now give organizations the ability to guide the brain.

The way you make decisions evolves, right? So context evolves. Am I going to be in full self-driving mode? Am I going to be in semi-automatic mode? Am I going to be in a manual mode? There's a very elegant way to say that, which comes from our friends at Unilever where they say, you have three modes with cognitive automation, human in the loop. So assisted human on the loop basically validating the recommendation.

So in the loop, you do the work with all the data and the capabilities provided by the engine or the brain. On the loop, you're interacting with the brain. Or out of the loop. You're literally out of the loop. The brain is doing the work. At that point, you need to be able to monitor in real time, the impact of the brain. How is the work going? What is the value that's being generated with this brain?

So it's a very good analogy. And today, I would tell you that we have clients, some of the largest companies in the world. You can check our cognitiveautomation.com. It's our community where customers talk about this stuff that have now moved from in the loop to, on the loop, to out of the loop. So it's now a reality, and I think it's catching up quite quickly. Whereas when you and I started talking about this a few years back, it was still a vision.

Brian Solis:

I love it. I love to see how far you've come and proud to have been part of those early days. John, I'm going to turn it over to you with the time that we have left.

John Kao:

Good. I have a mundane and a more metaphysical question. The mundane question is how does Aera show up in my enterprise? I mean, what's the user interface and experience? Is it is a web based experience? Is it a box that sits on my desk? I mean, just the nuts and bolts of how Aera shows up.

Frederic Laluyaux:

You access it through the browser on your desktop, but more interestingly, you access it with the voice or your mobile application. Why did we do that, John? Because the decisions that you're asked to make can be time sensitive. So just like Siri or Alexa, you need to be able to say, turn on the lights now, not tomorrow. And we wanted to bring that engagement wherever you are, whenever you need to engage.

So you can get in your car in the morning and say, Aera, what are the top five recommendations that I need to look at? And it will say, John, I have two recommendations on demand forecasting, one on pricing optimization, and one there. What are these? So you interact with the voice, or interact with the mobile, or you interact with any enterprise software through your desktop.

John Kao:

And do you see this also as B2C as well as a B2B play, in other words?

Frederic Laluyaux:

Yeah, you could conceptually think about it, but we haven't touched that. There are a few big stones that we had to turn to make this possible. The first one is the data, right? We spent years working on that because if you think about digitizing... What humans, John are really good at is approximation. We are having a dialogue. You don't have 100% of the data, but you have experienced tribal knowledge and you say I take a call, I'm going to make these decisions.

Now, when you do that, you introduce biases, and that's a problem. It's a problem with AI in general, but it's also a problem for the efficiency of the decision that you're supposed to make. And when you put that at scale, it creates a lot of inefficiencies that have an impact on the environment and so on and so forth. So there's pros and cons there. For the brain to actually make a call, it cannot take approximation.

So with the problem that we had to resolve is 100% of the data, the information that is required for a decision to be digitized has to be available into a single data model that the brain can connect to and understand. And that's a very big deal when you think about dozens of ERP and billions of transactions. That's the first thing we had to fix. So our technology has been really anchored on that as opposed to B2C. But in general, I think you could think about it as-

John Kao:

Theoretically. I mean, it seems like what you're offering is valuable to a variety of end users, and the enterprises may be the big concentrated customers, but there are other applications as well. Which leads me to the next question, which is how do you get the arms and legs to connect with the brand?

I mean, so the brand is doing the cognitive processing, but now you're kind of linking it to your accounting system or you're linking it to inventory management or if it's in my home, it's home automation or it's my wellness envelope of data turned into recommendations. I mean, how do the arms and legs grow out of the brain?

Frederic Laluyaux:

Yeah. I love this. In the early days, we talked about the analogy to explain what we do with the self-driving car. And we said, look, what you're doing in self-driving car is... What Tesla and a few others have done, they've digitized the operating system of the car. And what is the operating system of the car? It's the driver. And what does the driver do? It captures all the signals and analyzes them.

And that can be short-term decision, press on the brake, or long-term decision, I'm going to take this freeway versus that freeway in order to get to my destination. Just saw Wendy popping up here. This very nice. Hello, Wendy. Always good to see you. So the brain then, once you make a recommendation and you make a decision, then you're absolutely right. Connect with my arms and my legs to actually move the car.

So this is why it's so difficult to retrofit an existing car with a digital brain. You can't do it because that connection doesn't exist, right? So their two characteristics is A, all the components of the brain that allow a decision to be made have to be aware of one another, and it's very difficult to do, right? So memory needs to connect with the brain back into the memory. And there's a lot of work here. But the second big touchstone is connecting to the nervous system.

So without getting too technical, John, we built a technology that's quite similar to what Google did with the crawlers, when they crawl the Internet and they basically build a real-time replica of your website on their environment and then they were able to index it and do a lot of work to connect it to their brain. We've taken a very similar analogy. We're able to connect to those ERPs and those planning tools and those transportation systems in real time thanks to a potential technology that we have that we call the crawlers.

John Kao:

So here comes the metaphysical question and you're probably, or you may be familiar with a new book by Daniel Kahneman called Noise and how humans exhibit all kinds of biases in their decision-making reason. And you're offering kind of a more apparently objective set of solutions. However, I guess my question is how are you going to sort out the different levels of utility of having this capability?

Because for instance, you said something about you have to work out the issue of trust between the technology and humans. Well to me, trust building is something that's inherently about what humans do. It's I am trustworthy because a human perceives me as being trustworthy because of my human behaviors. And I would view the trustworthiness of Siri, for example, as being more a question of reliability.

So I can count on Siri, but I don't necessarily trust Siri in the sense that we can have now an intimate relationship, or they're going to empathize with me or understand me and so forth and so on. So if we then take that back to what tasks your technology will assume, it's a no brainer to make certain things more efficient. I mean, to reduce the noise in a supply chain or something like that.

Where I think the open questions lie, and I'm curious to know where you think the frontiers might be, is where I guess a more qualitative kind of judgment begins to emerge, which is the kind of understanding based on the understanding. So it's no longer about simply manipulation of data, but it's about pattern recognition within that data, it's about higher order interpretation of the data, and it's also about having a certain humanoid, I won't say human, level of empathy as to what the end users really need.

So when you talk about I never want to have mosquito bites again, the AI doesn't go out and kill everybody because that means there'll be no more mosquito bites. They have an understanding of the empathy for the human dimension. So that's probably three big questions rolled into one. But you got me thinking here.

Frederic Laluyaux:

And you got me thinking back. I think the word trust, you're right. It's trust on the reliability, right? I mean, it's not trust at the human level, but it's trust on the reliability. Can I trust the recommendation that is made? Can I trust the fact that the logic that is applied and the data that is feeding that recommendation that will lead to a decision is reliable, right? This is just a basic stuff.

But that is what's polluting and preventing companies to gain scale. So if you're a resolve that problem, you actually enable a lot of good decisions to be made in real time. I want to bring this to the level of pragmatism of that you're right, the word trust can involve more than what I really meant. The second is can I let the system run on its own implies have I established the right guardrails?

Have I established the right boundaries? Do I feel that the boundaries have been set properly? And can I modify them over time? The systems are not that smart. We talk about intelligence. It's not that smart. It's just basically executing and deploying a massive scale some compute that is programmed and driven by people. So when we look at Aera, we always talk about going from people doing the work or making the decisions guided by machines, to machines doing the work, making the recommendations, but guided by people.

And the guided by people is incredibly important because your business context evolves very quickly, and you want to be able to reassign or realign the way the brain behaves in real time. And we've seen that during COVID. Some of the processes that were running in self-driving mode, literally hands off, the system is delivering the forecast for you, and the accuracy of the forecast is beating the humans over and over, and we feel really good about this.

Well, Brian, you talked about March last year. When your clients are selling PPE or gel to clean your hands, you can imagine that all the data signals are completely out of work, and you need to be able to very quickly take control and say, let me drive again. So all of the things that we're talking about automation, documentation, work, but we also provide the capabilities to monitor, to guide the way the brain is actually working.

And even the world brain is excessive in a way, right? It's a big computer. And that's why I talk about machines and computers. And I think the analogy with brain is quite tricky. But there is some intelligence. There is some ability to do some work cause analysis dynamically, there is AutoML and all those capabilities that will, for example, tell you and recommend to you that in the given context, I suggest that...

I don't want to have to go to a human operator for validation because I've got enough certainty that the effort to ask the human instead of system question is not worth it. And that's when it gets a bit tricky, right? It depends on how you want to let go or not.

John Kao:

Well, we're sort of running out of time. I'd love to explore maybe in another conversation what's your R&D agenda is because I could imagine a lot of interesting experiments that you could conduct including bringing in some potential opportunity owners that would say let's look at making my enterprise an AI data-driven enterprise that would give you use cases that may not be obvious right now. So I'll leave that as a little tantalizing bit of unfinished business and turn it back over to Brian.

Frederic Laluyaux:

Would love that.

Brian Solis:

Even on that note, John, even as a catalyst to become data-driven and connecting the dots across the board, which would unfold and unlock all kinds of opportunities across the enterprise as well. Well, Fred, it's such a pleasure. Such a pleasure. This is exciting. I was-

John Kao:

Congratulations.

Brian Solis:

Yes. Congratulations. Very much looking forward to this, and I do look forward to seeing you in San Francisco for a glass of wine.

Frederic Laluyaux:

I hope we can do that too. We can do that very soon. And thanks for having me. It's always a real pleasure and got me thinking, guys. That's what I need.

John Kao:

Sounds dangerous.

Brian Solis:

Excellent, Fred. Talk to you soon, my friend.

Frederic Laluyaux:

Thank you.

Brian Solis:

Oh man.

John Kao:

Sound good.

Brian Solis:

He's just such a good person. So smart. And he's already had successful exits and just a good person. Well, John, all right.

John Kao:

All right. Here we are again.

Brian Solis:

Another one. And hey, the work happening behind the scenes wasn't too bad.

John Kao:

Well, I muted the loudest parts because they're drilling into concrete to put some mounts. But it worked out okay. I mean, I think in a way it was better that I wasn't watching and agonizing over it because they got to do their work without interference from me.

Brian Solis:

Wonderful. Well, everyone, thank you so much for tuning in. This has been an exceptional episode. I'm fired up on a lot of fronts as always. My brain continues to expand in wonderful ways. John, I always enjoy our time together. Thank you. And for those who are watching, thank you. Please do tune in each week, Thursdays 10:30 AM Pacific. Do comment in all of these threads.

Brian Solis:

We do check them during the week. And also to view past episodes because I don't think anybody that we've had on yet is irrelevant in terms of the conversations we've had and the state of the world and where it needs to go. Please go to intersectionslive.com and indulge. That's why we do this every week. So with that said, Gregarious, my friend, if you can take us out with the outro. We'll see you next week.

John Kao:

See you all next week.


By
From The Editor
,
Published:
September 10, 2021
Share:
SUBSCRIBE TO WEEKLY EMAIL
-->