Enterprise AI That Manufactures Results

Connected intelligence that keeps manufacturing ahead of disruptions, downtime, and demand shifts.

Proven performance at every stage of manufacturing.

Resilient Supply

40–65%

Reduction in critical stockouts

Reliable Assets

20–50%

Reduction in downtime

Optimized Production

1–2%

Increase in yield

Built for Where Manufacturing Breaks Down.

Supply Disruptions

Volatile demand, global sourcing complexity, and limited supplier visibility make it difficult to align inventory with production. Excess stock ties up capital, while shortages disrupt production lines.

dehaze

The C3 AI DifferenceSupply Resilience

Supply chain optimization detects and responds to disruptions before they impact the production floor.

dehaze

Unplanned Downtime

One failed machine can stop an entire production line. Across complex, interdependent asset fleets, unplanned downtime is a constant risk without early warning.

dehaze

The C3 AI DifferencePredictive Maintenance

AI-driven predictive maintenance continuously monitors asset health across the fleet, surfacing risks and recommended actions before they become production stops.

dehaze

Production Inefficiency 


Fixed setpoints and manual tuning can’t keep pace with live conditions. Deviations in temperature, pressure, and feed rates compound into lost yield, quality defects, and wasted energy.

dehaze

The C3 AI DifferenceProcess Optimization

Machine learning continuously adapts process setpoints to real conditions, catching deviations before they compound.

dehaze

Software for Intelligent Manufacturing

Purpose-built for the supply, reliability, and production demands of manufacturing.

C3 AI Demand Planning

Accurately forecast demand at any granularity and time horizon by unifying order history, operational data, and external signals, translating forecasts directly into executable production plans to reduce stockouts and improve service levels.

See the Features

This 9-step guided tour demonstrates how C3 AI Demand Planning helps organizations generate accurate forecasts and create unified demand plans. The tour covers AI-generated forecast accuracy and bias, transparent evidence packages with key financial drivers, what-if scenario modeling, AI-guided forecast evaluation, dynamic hierarchy reconciliation, and cross-functional demand plan approval and promotion. To request a personalized demo, complete the contact form at the end.

C3 AI Reliability

Predict equipment failures weeks before they occur. A unified view of sensor, maintenance, and operational data across every asset gives maintenance teams the early warning they need to act before failures disrupt production.

See the Features

This 9-step guided tour demonstrates how C3 AI Reliability helps organizations improve uptime and productivity. The tour covers high-priority KPI dashboards, ranked critical risk alerts, deep-dive alert analysis with key drivers and failure insights, AI-powered generative search for alert context, failure mode recommendations for root cause analysis, and direct work order creation to resolve issues quickly. To request a personalized demo, complete the contact form at the end of the tour.

C3 AI Process Optimization 

Optimize process conditions in near real time to improve yield, quality, and energy efficiency. By combining AI and physics-based models, manufacturers can close the gap between current performance and true operating potential.

See the Features

This 8-step guided tour demonstrates how C3 AI Process Optimization helps organizations optimize process performance. The tour covers high-priority production KPIs, real-time AI-driven process recommendations, transparent optimization model details, historical trends and constraint analysis, and direct operator actions to initiate process improvements. To request a personalized demo, complete the contact form at the end of the tour.

Proven Results Across the Manufacturing Industry

Dow and C3 AI: Building the AI-Driven Manufacturing Enterprise

Dow started with C3 AI Reliability to maximize asset uptime across its global steam cracking furnace fleet and is now scaling AI as a foundational capability to drive efficiency and productivity enterprise-wide. Watch their Chief Information and Digital Officer, Debra Bauler, onstage at C3 Transform 2026.

Transcript

00:00:15,660 --> 00:00:19,140

Well Lord Sarfraz, thank you for the magnanimous


00:00:19,620 --> 00:00:24,585

introduction There. Really appreciate it. All right, So


00:00:25,545 --> 00:00:29,385

first of all, welcome, Deb. Really appreciate you joining us


00:00:29,685 --> 00:00:34,905

here. We started the morning with Jim Snabe, basically


00:00:34,905 --> 00:00:41,145

painting a great vision of the autonomous enterprise, And that


00:00:41,145 --> 00:00:46,890

was followed, not the five year old English teacher, that was


00:00:46,890 --> 00:00:52,170

followed by Nikhil, taking us through a whirlwind of the


8

00:00:52,170 --> 00:00:56,850

technology that actually enables that, And I would say, Deb, now


9

00:00:57,150 --> 00:01:00,810

this is the hard part, which is, how do you operationalize exactly


10

00:01:01,410 --> 00:01:05,550

in a massive Fortune 500 company, So thank you for


11

00:01:05,550 --> 00:01:11,475

joining us to kind of tell us how to do this. Maybe we can


12

00:01:11,475 --> 00:01:18,975

start with Dow and the complexity of operating a


13

00:01:18,975 --> 00:01:24,855

business like Dow and just paint the picture for us as a current


14

00:01:24,855 --> 00:01:25,275

state.


15

00:01:25,335 --> 00:01:27,915

Yeah, well, before I do that, I want to thank you


16

00:01:27,975 --> 00:01:31,275

for inviting us to be a part of today. It is a great honor to be


17

00:01:31,275 --> 00:01:35,400

here and amongst such great colleagues. When I think about


18

00:01:35,400 --> 00:01:37,920

like the purpose of coming, I just want to thank you and the


19

00:01:37,980 --> 00:01:40,560

entire team at C3 AI for getting us together, because it is


20

00:01:40,560 --> 00:01:43,020

really valuable to share the stories with each other and hear


21

00:01:43,020 --> 00:01:44,100

where you're going, So thank you.


22

00:01:44,160 --> 00:01:44,520

Super.


23

00:01:44,700 --> 00:01:47,940

Yeah, you know, let, it again. Thank you for


24

00:01:47,940 --> 00:01:51,180

the great intro about Dow. I feel like, you know, magnanimous


25

00:01:51,180 --> 00:01:54,840

is quite nice, how you put that, But for those who may not know


26

00:01:54,840 --> 00:01:58,905

who Dow really is, at the end of the day, you know, we are 130


27

00:01:58,905 --> 00:02:02,505

year old company, so we kind of have some legacy about how we


28

00:02:02,505 --> 00:02:07,425

work. We are a material science company. We operate with about


29

00:02:07,425 --> 00:02:11,985

100 different locations that you can think about. We have 35,000


30

00:02:11,985 --> 00:02:15,765

employees around the world, and we operate some of the world's


31

00:02:15,765 --> 00:02:20,250

most complex and energy intensive assets, And so when I


32

00:02:20,250 --> 00:02:23,490

think about that, you know, we then transform chemistry into


33

00:02:23,490 --> 00:02:26,730

great products. It is true, you are surrounded by Dow from the


34

00:02:26,730 --> 00:02:29,790

chairs we sit in, to the lipstick on my face, to the


35

00:02:29,790 --> 00:02:33,030

shoes we're wearing, to the clothes we have on, but you also


36

00:02:33,030 --> 00:02:35,610

to think about some of the fun things that we do, Right? I


37

00:02:35,610 --> 00:02:39,570

think a lot about this room might care about data centers


38

00:02:39,570 --> 00:02:43,215

and we think about moving from water cooling to liquid cooling


39

00:02:43,215 --> 00:02:47,475

to the silicone that might be on the chipset or in your phone.


40

00:02:47,715 --> 00:02:51,675

So you can think about us in a lot of different ways, but all


41

00:02:51,675 --> 00:02:56,175

that complexity ties back to our really big assets, And I love to


42

00:02:56,175 --> 00:02:59,535

tell people these fun facts. One of them alone is like the size


43

00:02:59,535 --> 00:03:04,035

of Manhattan, right? three miles by three miles is a facility, So


44

00:03:04,035 --> 00:03:07,320

now you think about that asset and that asset, How do we take


45

00:03:07,320 --> 00:03:10,200

care of it? And that is where some of our partnership comes


46

00:03:10,200 --> 00:03:13,800

into play, But those are really important things to us, to keep


47

00:03:13,800 --> 00:03:16,560

running, to serve all of our customers and you around the


48

00:03:16,560 --> 00:03:16,920

world.


49

00:03:17,320 --> 00:03:21,520

So that's the core, and the upstream, if you will, is


50

00:03:22,660 --> 00:03:28,360

are these, I guess you call them steam crackers, so massively


51

00:03:28,360 --> 00:03:32,440

energy intensive, massively complicated to operate, units


52

00:03:32,440 --> 00:03:36,700

and when those things are operating well, everything


53

00:03:36,700 --> 00:03:40,300

downstream of that, all the products you produce into the


54

00:03:40,345 --> 00:03:43,405

various end markets are basically flowing through.


55

00:03:44,785 --> 00:03:50,365

That's actually where we started our journey, Deb, which is asset


56

00:03:50,665 --> 00:03:53,545

performance or asset utilization, So if you can


57

00:03:54,025 --> 00:03:58,585

improve the asset utilization by a percent makes a


58

00:03:58,645 --> 00:04:02,005

massive difference to Dow and obviously keeping it running


59

00:04:02,005 --> 00:04:04,750

but also improving the efficiency, So maybe you can


60

00:04:04,870 --> 00:04:08,830

just talk a little bit about the journey there.


61

00:04:08,950 --> 00:04:10,510

Yeah, well, and I think one of the great


62

00:04:10,510 --> 00:04:14,470

partnership elements between us and C3 is the knowledge you


63

00:04:14,470 --> 00:04:17,830

bring, and that is not easy to impress our team members with


64

00:04:17,830 --> 00:04:20,890

having extensive knowledge, So I think let's start there with the


65

00:04:20,890 --> 00:04:23,890

high caliber of people and the knowledge you brought to the


66

00:04:23,890 --> 00:04:27,115

table, but also this partnership word, and I do mean that


67

00:04:27,115 --> 00:04:30,475

seriously, in that within Dow, and you might imagine in an


68

00:04:30,475 --> 00:04:34,135

environment where I think the numbers we have, like 1,600


69

00:04:34,135 --> 00:04:38,635

phDs, you know, we have this real value of insight and


70

00:04:38,635 --> 00:04:42,955

knowledge and s of, you know, engineers, when we can put


71

00:04:42,955 --> 00:04:45,355

those two together, And I think that's really been the nice


72

00:04:45,355 --> 00:04:49,660

partnership. Is we've been able to take our knowledge and our IP


73

00:04:49,660 --> 00:04:53,320

and our internal capabilities and models and match them with


74

00:04:53,320 --> 00:04:57,100

C3 and the platform to really focus and in this particular


75

00:04:57,100 --> 00:04:59,680

instance, we're talking about our asset and increasing asset


76

00:04:59,680 --> 00:05:02,860

reliability, And so working together wasn't flawless to


77

00:05:02,860 --> 00:05:04,900

begin with, right? As most things you work


78

00:05:04,900 --> 00:05:08,560

through and tweak these things, but I think we've been really


79

00:05:08,560 --> 00:05:11,440

pleased with the partnership, And then, of course, we started


80

00:05:11,440 --> 00:05:14,965

in a certain class, and now we're also extending beyond, So


81

00:05:14,965 --> 00:05:17,665

starting about very narrow, and that was our approach. I know


82

00:05:17,665 --> 00:05:20,245

different companies have a different approach, but right


83

00:05:20,245 --> 00:05:24,205

we try to be very narrow within the landscape of Dow to prove


84

00:05:24,265 --> 00:05:27,685

value and then extend to a broader set of our assets.


85

00:05:27,805 --> 00:05:33,685

And I think the point about your PhDs and data


86

00:05:33,685 --> 00:05:38,470

sciences, et cetera, I think Dow is now self sufficient at


87

00:05:38,470 --> 00:05:43,570

basically configuring models, tuning models, deploying models


88

00:05:43,570 --> 00:05:48,670

so that actually was, it started with C3 doing some of that work


89

00:05:48,670 --> 00:05:52,810

on and partnering with the domain expertise that Dow has on


90

00:05:52,810 --> 00:05:57,010

operating steam crackers and furnaces and all that, But very


91

00:05:57,010 --> 00:06:02,275

quickly, Dow became self sufficient at actually using C3


92

00:06:02,275 --> 00:06:06,235

to tune their own models, and so that was an important part of


93

00:06:06,295 --> 00:06:07,735

the consideration, right.


94

00:06:07,915 --> 00:06:10,675

For us, for sure, right, is to be independent to


95

00:06:11,095 --> 00:06:13,555

well, we're good at asset operations, right? So this is


96

00:06:13,555 --> 00:06:17,095

not too dissimilar. It's an asset that we want to operate.


97

00:06:17,635 --> 00:06:19,855

But you know, we're also having conversations, are there other


98

00:06:19,855 --> 00:06:24,160

places to use, in this case, the platform or an asset to us to


99

00:06:24,160 --> 00:06:26,500

broaden across the landscape of Dow and get more value.


100

00:06:27,280 --> 00:06:30,460

And the journey, I think, was actually, is very


101

00:06:30,460 --> 00:06:35,320

similar to what Nikhil presented, that Koch FHR is on


102

00:06:35,320 --> 00:06:41,440

which is to it's hard to use, but it's autonomous site


103

00:06:41,440 --> 00:06:46,705

operations, But a very obviously, we have to be very


104

00:06:46,705 --> 00:06:49,105

concerned about, extremely concerned about the safety of


105

00:06:49,105 --> 00:06:52,525

the operations. Autonomous is not a light word to use in the


106

00:06:52,525 --> 00:06:53,485

context of Dow.


107

00:06:53,060 --> 00:06:55,640

No, and you know, hey, if we did have to have


108

00:06:55,640 --> 00:06:58,640

people operating, there'd be great thing if they were fully


109

00:06:58,760 --> 00:07:01,940

safe and secure, But I think we have to wait some time and


110

00:07:01,940 --> 00:07:04,460

distance before we're going to really be in that situation


111

00:07:04,460 --> 00:07:06,920

today where we're not going to have people there to oversee the


112

00:07:06,920 --> 00:07:11,480

decisions, or at least surround So for sure, I love the idea of


113

00:07:11,480 --> 00:07:14,360

fully autonomous, There just has to be a lot of proof points on


114

00:07:14,360 --> 00:07:15,680

the way for us to get there.


115

00:07:15,740 --> 00:07:19,625

So it's like the assisted lane versus completely


116

00:07:19,625 --> 00:07:20,105

autonomous.


117

00:07:20,105 --> 00:07:22,025

It's somewhere between, I think it's somewhere


118

00:07:22,025 --> 00:07:25,025

between gyms two and five, right? I think we're in the


119

00:07:25,025 --> 00:07:28,565

assisted but we're advancing a bit, and our aspiration is


120

00:07:28,565 --> 00:07:29,705

probably somewhere in there.


121

00:07:29,705 --> 00:07:35,645

Got it, Super, So Dow announced this massive


122

00:07:35,645 --> 00:07:40,805

transformation, which is super impressive. I think it's called


123

00:07:40,850 --> 00:07:45,590

transform to Outperform. Could you give some context on that


124

00:07:45,710 --> 00:07:49,430

and actually the choice of Transform to Outperform as the


125

00:07:49,430 --> 00:07:50,030

title for it?


126

00:07:50,540 --> 00:07:53,240

Yeah, And so maybe, similar to some of the


127

00:07:53,240 --> 00:07:58,280

people in the room that we have had three years of being in the


128

00:07:58,280 --> 00:08:02,540

bottom of a cycle, and that has proven to be financially


129

00:08:02,540 --> 00:08:06,260

difficult, and we are underperforming for where we


130

00:08:06,260 --> 00:08:10,400

would like to be, So just having said that, you know, we are also


131

00:08:10,400 --> 00:08:14,360

looking really internally at ourselves and saying, okay, like


132

00:08:14,405 --> 00:08:17,945

we have got to transform the way we work. There's all sorts of


133

00:08:17,945 --> 00:08:21,125

capabilities that we should be embracing. As I mentioned, we're


134

00:08:21,125 --> 00:08:24,785

130 year old company. We might have some legacy ways of working


135

00:08:25,085 --> 00:08:29,525

that we need to challenge constructively and really try to


136

00:08:29,525 --> 00:08:32,825

do differently, So our ambition is to take out 2 billion of


137

00:08:32,825 --> 00:08:35,742

cost. 2 billion. Yeah, 2 billion is the number, And so when we


138

00:08:35,742 --> 00:08:36,329

think about that, our aspirations are going to come


139

00:08:36,329 --> 00:08:40,370

from growth, but there's also some aspirations that are going


140

00:08:40,370 --> 00:08:45,050

to come internally, And so internally it's going to be, you


141

00:08:45,050 --> 00:08:47,270

know, how do we get more efficient? how do we streamline?


142

00:08:47,270 --> 00:08:50,330

How do we really look at the way we work? and then, of course


143

00:08:50,330 --> 00:08:53,390

when we do that, AI and automation are a portion of it.


144

00:08:53,390 --> 00:08:55,790

I'm not quite sure how the media wants to cover that, but it


145

00:08:55,790 --> 00:08:58,550

seems sometimes there's a there's a higher correlation


146

00:08:58,550 --> 00:09:02,375

that AI is solving all those issues. It's not, but it is an


147

00:09:02,375 --> 00:09:06,095

important ingredient in the steps forward for us, But really


148

00:09:06,095 --> 00:09:09,455

transform to outperform is what it is, right? Like, if we can


149

00:09:09,455 --> 00:09:12,455

transform internally, then we're going to outperform the


150

00:09:12,455 --> 00:09:14,915

competition, and at the end of the day, that's what we want to


151

00:09:14,915 --> 00:09:17,975

do. We want to beat the competition to satisfy you, our


152

00:09:17,975 --> 00:09:19,955

customers, at the end of the day.


153

00:09:20,195 --> 00:09:23,720

So give an example or two of that, So it's operating


154

00:09:23,720 --> 00:09:28,040

more nimbly in the customer side and then the operation side. Can


155

00:09:28,040 --> 00:09:29,420

you bring it to life for us?


156

00:09:29,000 --> 00:09:31,280

I can bring that to life for you, And if you


157

00:09:31,280 --> 00:09:35,180

allow me to talk about, maybe I'll intertwine AI a moment for


158

00:09:35,180 --> 00:09:38,000

you, Right? If you allow me to do that, because it isn't wholly


159

00:09:38,000 --> 00:09:40,880

going to be AI, but I do want to at least share a bit about this.


160

00:09:40,880 --> 00:09:43,160

So when we think about, or at least I try to talk about AI


161

00:09:43,160 --> 00:09:46,280

internally. I try to talk about in three ways. One is you as a


162

00:09:46,280 --> 00:09:49,160

person, right? You have your own AI. You're gonna have your own


163

00:09:49,160 --> 00:09:52,760

AI agents, you know, hopefully you're all loving copilot in


164

00:09:52,760 --> 00:09:56,165

your daily lives. Our leaders love it to help at year end


165

00:09:56,165 --> 00:09:58,985

review time, right? Like it's certainly making you more


166

00:09:58,985 --> 00:10:03,185

efficient, And then the other two ways that I would be


167

00:10:03,185 --> 00:10:05,585

thinking about AI, at least within the landscape of Dow, is


168

00:10:05,585 --> 00:10:09,365

from our end to end processes, So you can think of your standard


169

00:10:09,365 --> 00:10:13,445

order to cash processes, right when we think about like, how do


170

00:10:13,445 --> 00:10:16,850

we have the start and the end, hire to retire, any of these end


171

00:10:16,850 --> 00:10:19,430

to end processes, we are starting to look at, how do we


172

00:10:19,430 --> 00:10:24,230

use AI agents and other automation more extensively than


173

00:10:24,230 --> 00:10:28,010

we have historically. The third way I would talk about AI, and


174

00:10:28,010 --> 00:10:32,810

do talk about it, is within specific domains, So within R&D


175

00:10:32,810 --> 00:10:37,070

how are we using AI to help us get to innovations faster? But


176

00:10:37,070 --> 00:10:39,995

that doesn't really affect the finance team, right? It's really


177

00:10:39,995 --> 00:10:42,815

contained within a domain, and this is where we often talk


178

00:10:42,815 --> 00:10:45,935

about C3 today for us, is playing in the domain of our


179

00:10:45,935 --> 00:10:49,535

operations space. You are helping us make our assets more


180

00:10:49,535 --> 00:10:52,475

efficient, and we're using A in that regard, but to stitch that


181

00:10:52,475 --> 00:10:55,175

all together in transform to outperform we're really


182

00:10:55,175 --> 00:10:59,375

evaluating end to end, and those processes and practices in


183

00:10:59,375 --> 00:11:04,220

streamlining the way we do the work we do, and where will AI be


184

00:11:04,220 --> 00:11:06,800

A big part of that, or just simply streamlining the


185

00:11:06,800 --> 00:11:10,460

processes we have? Supply chain planning might be a good place


186

00:11:10,460 --> 00:11:14,660

to think about in terms of streamlining, to transform the


187

00:11:14,660 --> 00:11:17,780

way we work and outperform where we are today and where our


188

00:11:17,780 --> 00:11:18,680

competition is.


189

00:11:18,680 --> 00:11:21,860

Super, I like that framework. It actually mirrors a


190

00:11:21,980 --> 00:11:24,500

little bit of what Nikhil was Does.


191

00:11:24,560 --> 00:11:28,145

It Does. That was not planned. That was not


192

00:11:28,445 --> 00:11:28,925

planned.


193

00:11:29,405 --> 00:11:34,385

So I think Dow, you've identified an order of nine end


194

00:11:34,385 --> 00:11:35,405

to end processes.


195

00:11:35,405 --> 00:11:38,045

Yeah, nine or ten, somewhere in the direction


196

00:11:37,400 --> 00:11:41,662

and then for each of those there might be


197

00:11:38,105 --> 00:11:38,285

yeah.


198

00:11:41,751 --> 00:11:46,013

Supporting domain specific capabilities that get


199

00:11:46,102 --> 00:11:51,430

orchestrated, And then this team level automation, that's


200

00:11:51,519 --> 00:11:56,936

kind of the third layer in this process, And so when you pull


201

00:11:57,025 --> 00:12:01,909

all this together, Dow transforms, so you basically are


202

00:12:01,998 --> 00:12:07,326

forecasting better what products are needed, so that you can


203

00:12:05,165 --> 00:12:08,765

exactly you've got it.


204

00:12:07,415 --> 00:12:12,566

Schedule production better, so that you can have the right


205

00:12:10,805 --> 00:12:16,565

that's the picture.


206

00:12:12,654 --> 00:12:18,160

Levels of inventory and so you can serve the customer, on time


207

00:12:16,805 --> 00:12:25,550

in full Exactly you've got it That's the picture. Maybe if this is relatable to


208

00:12:18,249 --> 00:12:18,960

in full


209

00:12:25,550 --> 00:12:28,670

some, but within the landscape of Dow, we have some businesses


210

00:12:28,670 --> 00:12:32,030

that are truly more of a commodity type business, and


211

00:12:32,030 --> 00:12:34,910

then we have a specialty type business. You'd know that those


212

00:12:34,910 --> 00:12:37,790

are different in terms of sophistication and complexity


213

00:12:37,790 --> 00:12:41,210

for sure, if you try to manage your working capital in a


214

00:12:41,210 --> 00:12:44,510

certain way to optimize what is good for your commodity


215

00:12:44,510 --> 00:12:46,850

business. It's actually not great for your specialty


216

00:12:46,850 --> 00:12:48,635

businesses, right? Because you're trying to reduce your


217

00:12:48,635 --> 00:12:52,355

working inventory, So we're trying to manage between these


218

00:12:52,655 --> 00:12:56,975

dualities that we live in terms of the landscape of Dow


219

00:12:57,275 --> 00:12:59,735

and for sure, to your point about trying to streamline these


220

00:12:59,735 --> 00:13:03,215

processes in terms of, can I have a little bit more inventory


221

00:13:03,215 --> 00:13:07,115

in these and how do I optimize for a specialty business, and


222

00:13:07,115 --> 00:13:10,070

where can I really get efficient on my commodity business in the


223

00:13:10,070 --> 00:13:12,860

inventory? So it is a struggle today. Like I said, we've got


224

00:13:12,860 --> 00:13:15,860

some legacy things that we're working on, but our aspiration


225

00:13:15,860 --> 00:13:19,280

is to really aggressively, and that's why we're announcing


226

00:13:19,280 --> 00:13:22,700

these things, is to aggressively hit at these different ways of


227

00:13:22,700 --> 00:13:23,060

working


228

00:13:23,240 --> 00:13:26,300

and it is aggressive. I mean, your timelines, this is 2


229

00:13:26,300 --> 00:13:29,660

billion, not in five years.


230

00:13:29,660 --> 00:13:32,840

No, it's in two years. Yeah, And we already had a


231

00:13:32,840 --> 00:13:35,360

brilliant. Like, we already had one before we tried to do


232

00:13:35,360 --> 00:13:36,500

you know, so, yeah.


233

00:13:36,560 --> 00:13:41,540

So this is moving fast, so maybe turn to, Deb


234

00:13:41,600 --> 00:13:45,020

the hard part, which is, how do you get everybody on the same


235

00:13:45,140 --> 00:13:48,725

page? How do you get it working with business, working with your


236

00:13:48,725 --> 00:13:52,085

commercial, working with, so, how do you operationalize this?


237

00:13:52,085 --> 00:13:55,445

Yeah, and, you know, it's not an easy thing, right? I


238

00:13:55,445 --> 00:13:58,445

think we've all would probably have some commonality in that. I


239

00:13:58,445 --> 00:14:01,385

think the number one thing to work on is, our collaboration


240

00:14:01,385 --> 00:14:03,485

right? and I think to where we started, even with our


241

00:14:03,485 --> 00:14:05,825

partnership, is you got to work through some of the difficult


242

00:14:05,825 --> 00:14:08,405

moments. Think we're all aligned on the prize. Sometimes it's how


243

00:14:08,405 --> 00:14:09,950

you get there, And so it's a really strong collaboration that


244

00:14:09,950 --> 00:14:13,730

we're trying to bring between the business and the outcomes


245

00:14:13,730 --> 00:14:17,270

they need. Our team as a technology team, where is our


246

00:14:17,270 --> 00:14:20,570

purpose in helping achieve those outcomes? And then we think


247

00:14:20,570 --> 00:14:24,050

about partners like C3, how do we have you be a part of the


248

00:14:24,050 --> 00:14:26,450

table? we're not going to build everything. I love again, the


249

00:14:26,450 --> 00:14:30,050

platform, And I think Nikhil's question is, Are you gonna go


250

00:14:30,050 --> 00:14:33,995

build all your own? no, we just don't have the time at our pace.


251

00:14:34,355 --> 00:14:36,755

How do we strategically think together? Where do we get to


252

00:14:36,755 --> 00:14:40,655

leverage a platform and a strong partnership to bring into that


253

00:14:40,655 --> 00:14:43,295

intimate circle of really trying to transform yourself, because


254

00:14:43,295 --> 00:14:45,935

you're not going to bring in every vendor to do that, but


255

00:14:45,935 --> 00:14:49,175

where we have strong vendors and strong partnerships, I think is


256

00:14:49,175 --> 00:14:53,015

really important to collaborate and work on the really difficult


257

00:14:53,255 --> 00:14:58,280

task at hand, with candor, with trust, and most importantly


258

00:14:58,280 --> 00:14:59,180

with the results.


259

00:14:59,780 --> 00:15:03,260

Super, and so just to kind of summarize that, where


260

00:15:03,260 --> 00:15:06,920

there are pre-built applications you can roll out very quickly


261

00:15:07,640 --> 00:15:13,160

that might be like, C3 AI Demand Forecasting or supply network


262

00:15:13,160 --> 00:15:19,400

risk, etc, For the long tail, there is C3 Code, which


263

00:15:19,400 --> 00:15:21,560

you saw this morning.


264

00:15:21,560 --> 00:15:24,500

I did get a personal demo this morning because I had to


265

00:15:24,500 --> 00:15:25,565

miss some things yesterday.


266

00:15:25,685 --> 00:15:26,585

What did you think?


267

00:15:26,900 --> 00:15:29,420

No, you know, actually, you know my team was part of my team was


268

00:15:29,960 --> 00:15:33,200

with me, and I was, like, doing a little elbow jab. Because I


269

00:15:33,200 --> 00:15:36,860

was like, okay, we have some really old code, not that any of


270

00:15:36,860 --> 00:15:39,740

you have this problem, we have some tech debt in some of our


271

00:15:39,740 --> 00:15:43,700

environments, including like VB6, which I know we're, you


272

00:15:43,700 --> 00:15:47,360

know, alone in that category, Everybody else has fixed that


273

00:15:47,360 --> 00:15:49,520

problem, But we were, we were at least saying, hey, that


274

00:15:49,520 --> 00:15:52,685

might be really transformative to taking care of some of our


275

00:15:52,685 --> 00:15:55,985

tech debt. We had been talking about supply chain and using


276

00:15:56,105 --> 00:16:00,605

capabilities, but with code, we were definitely brainstorming


277

00:16:00,605 --> 00:16:04,505

from that venue to this one, about where could we really see


278

00:16:05,285 --> 00:16:08,705

some capabilities and some value, and how could we go do a


279

00:16:08,705 --> 00:16:12,005

pilot on that pretty quickly, And I think everybody's pretty


280

00:16:12,005 --> 00:16:14,450

optimistic that there's some real value to be had in the tech


281

00:16:14,450 --> 00:16:18,350

debt and remediation of tech debt without maybe as much


282

00:16:18,350 --> 00:16:20,690

effort, because we just don't have the time and energy to


283

00:16:20,690 --> 00:16:22,550

replatform all of those things, Right?


284

00:16:22,550 --> 00:16:24,650

I actually hadn't thought about that angle, but


285

00:16:25,250 --> 00:16:29,810

that's true. As Nikhil kind of showed again, if you can express


286

00:16:29,810 --> 00:16:35,270

the business problem in English, we can then interpret that and


287

00:16:35,270 --> 00:16:38,795

generate code that's consistent to the platform, So yeah, it's


288

00:16:38,795 --> 00:16:42,755

using all the same ontologies. It's using all the same security


289

00:16:42,815 --> 00:16:48,155

guardrails and interaction with the LLMs, etc, to actually


290

00:16:48,155 --> 00:16:52,655

recreate and reduce the footprint. Of a lot of it


291

00:16:52,655 --> 00:16:53,735

costs in that area.


292

00:16:54,215 --> 00:16:58,415

Again, and very old assets that are all around the


293

00:16:58,415 --> 00:17:02,780

world with different people who created different solutions over


294

00:17:03,080 --> 00:17:07,160

A long period of time, you know, it's just too much effort to try


295

00:17:07,160 --> 00:17:11,840

to harmonize those all in one big effort, And so it'll be


296

00:17:11,840 --> 00:17:16,040

interesting to see, like we love to be a proof point, at least in


297

00:17:16,040 --> 00:17:20,240

our brainstorm on the way from there to here, which was, hey


298

00:17:20,240 --> 00:17:25,205

could you help us with that, let alone the other tasks at hand?


299

00:17:25,205 --> 00:17:27,965

Because most of most IT budgets, I'm not


300

00:17:27,965 --> 00:17:31,385

asking about yours here. Deb, just to be clear, is basically


301

00:17:32,105 --> 00:17:34,505

in maintenance and keeping the lights on.


302

00:17:34,000 --> 00:17:37,900

It is, And if you want to transform to outperform, how


303

00:17:37,900 --> 00:17:40,120

do you lower your cost of maintenance and increase the


304

00:17:40,120 --> 00:17:43,300

oxygen you need to do the new fun stuff, right? and I think


305

00:17:43,300 --> 00:17:47,860

classically, if you're running a technology budget, you're really


306

00:17:47,860 --> 00:17:52,180

good if you got 80% maintenance and 20% projects, wouldn't that


307

00:17:52,180 --> 00:17:54,820

be fun if that were a little bit different? I mean, I'd love to


308

00:17:54,820 --> 00:17:58,300

tip that upside down, But even if I got to 50/50, in a


309

00:17:58,300 --> 00:18:01,225

manufacturing world now, financials and banks have


310

00:18:01,225 --> 00:18:04,465

different metrics for that 80/20 rule, but in our landscape, if


311

00:18:04,465 --> 00:18:08,365

we could really pivot that, you could either create the oxygen


312

00:18:08,365 --> 00:18:11,305

you need to go do the new things, or lower your overall


313

00:18:11,305 --> 00:18:11,785

cost.


314

00:18:11,785 --> 00:18:13,525

Super, something we could talk about


315

00:18:13,585 --> 00:18:14,245

we will.


316

00:18:14,785 --> 00:18:15,565

Off stage.


317

00:18:17,125 --> 00:18:19,465

Have the team join in, right? Have everybody join


318

00:18:19,465 --> 00:18:19,585

in.


319

00:18:19,765 --> 00:18:24,370

Let's do it. Just a couple of couple of other


320

00:18:24,370 --> 00:18:28,390

topics. I know Microsoft is a huge partner for Dow.


321

00:18:28,390 --> 00:18:29,050

Absolutely.


322

00:18:29,230 --> 00:18:32,650

Can you talk a little bit about how you see kind of C3


323

00:18:32,650 --> 00:18:38,350

and Microsoft accelerating Dow's journey?


324

00:18:39,250 --> 00:18:43,043

Yeah, And I think this is what I meant earlier


325

00:18:43,043 --> 00:18:43,064

when I was talking about the collaboration we need to


326

00:18:43,064 --> 00:18:45,895

transform. I think we need those few really strong partners, And


327

00:18:45,895 --> 00:18:49,075

you know, trust is earned. I don't know that it's always just


328

00:18:49,075 --> 00:18:54,295

granted, And when I think about the partnership, C3 AI and


329

00:18:54,295 --> 00:18:57,175

microsoft come with already a great partnership, and we


330

00:18:57,175 --> 00:18:59,695

individually have great partnerships, So I think, you


331

00:18:59,695 --> 00:19:06,055

know, we're better together when we can come in and deal with the


332

00:19:06,055 --> 00:19:06,081

difficult situation, So Microsoft's just been a


333

00:19:06,081 --> 00:19:07,120

tremendous asset along the way. You know, in our Azure


334

00:19:07,120 --> 00:19:11,680

environment, we're exclusive to Azure in that regard, So we do


335

00:19:11,680 --> 00:19:15,040

have high expectations, though, that for both of those


336

00:19:15,040 --> 00:19:18,760

companies, you bring us your best resources, And so I do


337

00:19:18,760 --> 00:19:21,160

think, as I said earlier on, like you have brought and


338

00:19:21,160 --> 00:19:24,220

impressed our organization with some really great talent, And I


339

00:19:24,220 --> 00:19:26,620

think the combination of Microsoft's talent, Dow's


340

00:19:26,620 --> 00:19:30,865

talent, and C3 AI's talent, I get excited about, what else can


341

00:19:30,865 --> 00:19:33,145

I challenge you guys with, or what else can we come up with


342

00:19:33,565 --> 00:19:34,285

along the way?


343

00:19:34,660 --> 00:19:38,080

Yeah, and C3 is obviously working with Microsoft


344

00:19:38,080 --> 00:19:41,920

to integrate and use all their latest services, And then we're


345

00:19:41,920 --> 00:19:45,220

working with Dow to make that, operationalize that adapt.


346

00:19:45,580 --> 00:19:49,960

There is one other kind of dimension, Deb, that I wanted to


347

00:19:49,960 --> 00:19:56,440

touch on, which is, you have a innovation arm, which is a


348

00:19:56,440 --> 00:19:58,405

licensed technology licensing arm.


349

00:19:58,405 --> 00:19:58,825

We do.


350

00:19:59,965 --> 00:20:03,985

Maybe talk a little bit, just give the audience a


351

00:20:03,985 --> 00:20:08,725

little bit of context on that, because I think you package up


352

00:20:08,725 --> 00:20:13,705

your IP and, yeah, show others how to run steam crackers across


353

00:20:13,705 --> 00:20:14,245

the globe.


354

00:20:14,245 --> 00:20:16,945

Yeah, I'm gonna do a little bit of a talk about, I


355

00:20:16,945 --> 00:20:19,765

have some experts in the room, So after this, like, if you have


356

00:20:19,765 --> 00:20:22,150

interest in understanding innovation. There's a great team


357

00:20:22,150 --> 00:20:26,050

here who would be happy to engage more, But if I, at the


358

00:20:26,050 --> 00:20:29,230

beginning, told you about the various expertise we have


359

00:20:29,230 --> 00:20:33,910

between our engineering and our PhD chemists, we do in the


360

00:20:33,910 --> 00:20:38,290

traditional material science way, have abilities to license


361

00:20:38,410 --> 00:20:42,670

our IP and that's innovation, is the arm that licenses those


362

00:20:42,670 --> 00:20:47,155

things, and over time, we have partnered with C3 AI to say what


363

00:20:47,155 --> 00:20:52,255

we have created in this space for our assets and how we're


364

00:20:52,255 --> 00:20:57,475

actually using C3 AI along with our IP, Should this not be


365

00:20:57,475 --> 00:21:00,775

something that we can help others who run furnaces and


366

00:21:00,775 --> 00:21:05,395

crackers and assets like we do, jump start and leapfrog to a


367

00:21:05,395 --> 00:21:09,880

solution, So innovation is the way of using what we have


368

00:21:09,880 --> 00:21:13,600

co-created and licensing it to others so they can have speed to


369

00:21:13,600 --> 00:21:16,840

market, instead of having to start at ground zero and


370

00:21:16,840 --> 00:21:19,660

benefiting from our investment, but also we can pay it forward


371

00:21:19,660 --> 00:21:22,060

and then earn a little bit from that IP along the way.


372

00:21:22,240 --> 00:21:24,460

So that's a that's another dimension of the


373

00:21:24,460 --> 00:21:28,480

partnership, which is when we say, deploy C3 AI Asset


374

00:21:28,480 --> 00:21:33,505

performance or Reliability into Dow, we're actually


375

00:21:33,505 --> 00:21:37,345

incorporating Dows decoking- models, proprietary models, if


376

00:21:37,345 --> 00:21:41,125

you will, then that solution is packaged up and then made


377

00:21:41,125 --> 00:21:43,765

available to the industry.


378

00:21:43,765 --> 00:21:46,645

Yep made available to industry Unovation is happy to talk about that and what it


379

00:21:46,645 --> 00:21:50,785

might bring to help others, As I said, like accelerate from what


380

00:21:50,785 --> 00:21:52,825

we have spent time, because we've been working for a few


381

00:21:52,825 --> 00:21:55,030

years together, right? I think we've been four years at this.


382

00:21:55,030 --> 00:21:59,230

So how do we help you accelerate and get to value as well?


383

00:21:59,530 --> 00:22:04,810

Super, maybe I can just conclude on what advice would


384

00:22:04,810 --> 00:22:08,830

you have, Deb, for this audience that's embarking on their


385

00:22:08,830 --> 00:22:09,790

transformations?


386

00:22:10,800 --> 00:22:13,140

You know, and it's always an interesting question Is, what


387

00:22:13,140 --> 00:22:16,080

advice do you have? I would really start and, first of all


388

00:22:16,140 --> 00:22:19,980

like the intentions you're trying to achieve, Sometimes I


389

00:22:19,980 --> 00:22:24,360

think we approach different efforts like these


390

00:22:24,360 --> 00:22:27,240

which are, hey, I just have a short term task, and I need some


391

00:22:27,300 --> 00:22:30,600

sort of result, and I'm gonna go fast, and I'm gonna call that a


392

00:22:30,660 --> 00:22:34,185

transactional situation. I think when we've approached C3 AI


393

00:22:34,185 --> 00:22:37,245

it's been more about a longer term journey, And if that is


394

00:22:37,245 --> 00:22:40,545

where you're at, then How do you treat this as a partnership


395

00:22:40,545 --> 00:22:42,585

where you're trying to say, okay, I'm going to start here


396

00:22:42,585 --> 00:22:46,185

but with a platform like this, there's a lot of capability that


397

00:22:46,185 --> 00:22:49,485

you can have in your future, but staying focused, and I will say


398

00:22:49,485 --> 00:22:53,925

we stayed very focused in our asset way, right, in terms of


399

00:22:53,925 --> 00:22:57,150

looking at this is where we want to target and now through a


400

00:22:57,150 --> 00:22:59,850

partnership, So I guess my advice is, are you trying to


401

00:22:59,850 --> 00:23:06,510

achieve a short term tactical solution to fit a problem, or


402

00:23:06,510 --> 00:23:09,270

are you trying to achieve a longer term partnership with


403

00:23:09,270 --> 00:23:11,970

greater outcomes? And again, we've been able to co-create and


404

00:23:11,970 --> 00:23:14,910

collaborate and create a better outcome, because I think we


405

00:23:14,910 --> 00:23:19,335

oriented ourselves on commitment to partnership, and this is


406

00:23:19,335 --> 00:23:22,575

quite transformative, if you choose it to be, So I would


407

00:23:22,575 --> 00:23:25,215

orient myself on that, And then I think the other thing is, be


408

00:23:25,215 --> 00:23:28,095

open to new ways of working, like we've really had to explore


409

00:23:28,095 --> 00:23:31,095

different approaches, and a trusted partner can hold a


410

00:23:31,095 --> 00:23:33,495

mirror up and say, well, yeah, I know you've done it this way


411

00:23:33,495 --> 00:23:36,135

but perhaps we should think about doing it a different way.


412

00:23:36,135 --> 00:23:39,555

That's great. Thank you for that, but, it's really


413

00:23:39,555 --> 00:23:43,260

consistent with what we're focusing the company on, which is better


414

00:23:43,260 --> 00:23:49,140

supporting rapid scale up of these transformations, so it's


415

00:23:49,200 --> 00:23:53,040

absolutely consistent with that, We're really looking forward to


416

00:23:53,100 --> 00:23:58,680

going on this autonomous Dow journey with you, and really


417

00:23:58,740 --> 00:24:01,560

really appreciate you joining us today. Thank you.


418

00:24:01,620 --> 00:24:04,020

No, thanks for having me And hopefully everybody has a


419

00:24:04,020 --> 00:24:04,860

great transform.


420

00:24:04,860 --> 00:24:06,120

Excellent, Yeah thanks so much.


Holcim runs C3 AI Reliability on edge devices to bring predictive maintenance alerts closer to plant operations, helping teams act quickly on critical equipment issues.

Read the Blog

Holcim and C3 AI: Bringing Predictive Maintenance to the Industrial Edge

manufacturing_proofpoint-2.png

Nucor and C3 AI: Unlocking Cost Savings Across the Steel Value Chain

Nucor partners with C3 AI to deploy AI across its entire steel value chain, from raw materials and demand forecasting to production scheduling, improving forecast accuracy and driving multimillion-dollar savings across its operations. Hear from Gaurav Malhotra, Department Manager, AI & Emerging Technology at Nucor.

Additional Resources

Maximizing Every Fuel Cell at Bloom Energy
A fireside-chat-style conversation on stage at C3 Transform 2026. Two speakers sit in white lounge chairs with a small side table between them holding a water bottle. On the left, a man in a navy suit and glasses leans on his hand, listening; on the right, a woman in a dark dress gestures mid-sentence. The large white C3.ai logo and wordmark are displayed on a black screen above them, with teal stage lighting in the background.

Bloom Energy manufactures solid oxide fuel cells at gigawatt scale, and uses C3 AI Process Optimization to get more output and longer life out of every one. Hear from their CMO, Natalie Sunderland, onstage at C3 Transform 2026.

Transforming Refinery Operations at Flint Hills Resources
A speaker presenting at what C3 Transform 2026. He wears a dark blazer over a light blue checked button-down shirt with a clip-on lapel microphone, gesturing with both hands as he talks. The background is deep navy blue with partial white event signage visible in the upper left.

Flint Hills Resources is using C3 AI Reliability to standardize asset monitoring across thousands of critical assets, driving more efficient anomaly resolution. Hear from their team onstage at C3 Transform.

Turning Demand Volatility Into Production Margin
Farming

C3 AI helped a global food manufacturer improve daily demand forecasts, reduce production scheduling effort by 96%, and identify $30M in additional gross margin from increased order fill rates.

Start Running Your Business on C3 AI.