1
00:00:04,519 --> 00:00:12,719
Speaker 1: Welcome to tech Stuff, a production from iHeartRadio. This season

2
00:00:12,800 --> 00:00:16,280
Speaker 1: on smart Talks with IBM, Malcolm Gladwell and team are

3
00:00:16,320 --> 00:00:19,560
Speaker 1: diving into the transformative world of artificial intelligence with a

4
00:00:19,600 --> 00:00:24,200
Speaker 1: fresh perspective on the concept of open What does open

5
00:00:24,640 --> 00:00:28,000
Speaker 1: really mean in the context of AI. It can mean

6
00:00:28,080 --> 00:00:31,680
Speaker 1: open source code or open data, but it also encompasses

7
00:00:31,800 --> 00:00:36,120
Speaker 1: fostering an ecosystem of ideas, ensuring diverse perspectives are heard,

8
00:00:36,240 --> 00:00:40,000
Speaker 1: and enabling new levels of transparency. Join hosts from your

9
00:00:40,040 --> 00:00:43,560
Speaker 1: favorite Pushkin podcasts as they explore how openness and AI

10
00:00:43,880 --> 00:00:49,120
Speaker 1: is reshaping industries, driving innovation, and redefining what's possible. You'll

11
00:00:49,159 --> 00:00:52,440
Speaker 1: hear from industry experts and leaders about the implications and

12
00:00:52,479 --> 00:00:56,240
Speaker 1: possibilities of open AI, and of course, Malcolm Gladwell will

13
00:00:56,280 --> 00:00:58,280
Speaker 1: be there to guide you through the season with his

14
00:00:58,440 --> 00:01:01,920
Speaker 1: unique insights. Look out for new episodes of Smart Talks

15
00:01:02,000 --> 00:01:05,440
Speaker 1: every other week on the iHeartRadio app, Apple Podcasts, or

16
00:01:05,440 --> 00:01:09,000
Speaker 1: wherever you get your podcasts, and learn more at IBM

17
00:01:09,160 --> 00:01:12,440
Speaker 1: dot com slash smart Talks.

18
00:01:13,480 --> 00:01:16,880
Speaker 2: Hello, Hello, Welcome to Smart Talks with IBM, a podcast

19
00:01:17,160 --> 00:01:22,800
Speaker 2: from Pushkin Industries, iHeartRadio and IBM. I'm Malcolm Godwell, this season,

20
00:01:22,840 --> 00:01:25,800
Speaker 2: we're diving back into the world of artificial intelligence, but

21
00:01:25,880 --> 00:01:31,760
Speaker 2: with a focus on the powerful concept of open its possibilities, implications,

22
00:01:31,800 --> 00:01:36,479
Speaker 2: and misconceptions. On today's episode, our season finale, I'm joined

23
00:01:36,480 --> 00:01:40,640
Speaker 2: by Rick Lewis, the senior vice president of Infrastructure at IBM.

24
00:01:41,360 --> 00:01:44,840
Speaker 2: Rick has had a remarkable career focused around product innovation.

25
00:01:45,560 --> 00:01:48,520
Speaker 2: He was actually a few years into retirement when IBM

26
00:01:48,560 --> 00:01:54,000
Speaker 2: came calling with an opportunity he just couldn't turn down. Thankfully,

27
00:01:54,200 --> 00:01:57,520
Speaker 2: Rick came out of retirement and today he oversees a

28
00:01:57,600 --> 00:02:03,080
Speaker 2: vast portfolio from storage and software to global customer support operations,

29
00:02:03,480 --> 00:02:05,800
Speaker 2: and he's engaged in one of the key problems facing

30
00:02:05,840 --> 00:02:10,720
Speaker 2: companies today, an explosion of data. In talking with Rick,

31
00:02:10,800 --> 00:02:13,160
Speaker 2: I can see that this problem of having so much

32
00:02:13,280 --> 00:02:17,360
Speaker 2: data is also an incredible opportunity because if you're able

33
00:02:17,400 --> 00:02:20,079
Speaker 2: to leverage that data to get the most value out

34
00:02:20,080 --> 00:02:22,480
Speaker 2: of it, then you can use it to help bring

35
00:02:22,520 --> 00:02:26,280
Speaker 2: your business into the future. We talked about the serious

36
00:02:26,320 --> 00:02:29,160
Speaker 2: computing power needed to scale AI, as well as the

37
00:02:29,200 --> 00:02:33,400
Speaker 2: ways that infrastructure storage solutions can be essential to enabling

38
00:02:33,480 --> 00:02:38,000
Speaker 2: this new world of possibilities. It's a really great conversation

39
00:02:38,520 --> 00:02:44,520
Speaker 2: so let's get to it. I'm here with Rick Lewis.

40
00:02:44,880 --> 00:02:45,079
Speaker 3: Rick.

41
00:02:45,320 --> 00:02:49,560
Speaker 2: Welcome, Thank Here. We are in the IBM's New York

42
00:02:49,560 --> 00:02:53,040
Speaker 2: City headquarters at one Madison Avenue. I'm going to start

43
00:02:53,040 --> 00:02:54,959
Speaker 2: with you're a hardware guy.

44
00:02:55,120 --> 00:02:58,760
Speaker 3: I'm a hardware guy. I grew up doing hardware chip engineering.

45
00:02:59,200 --> 00:03:01,680
Speaker 3: But like I tell a lot of people, a chip

46
00:03:01,680 --> 00:03:05,280
Speaker 3: engineering project is actually a giant software project with a

47
00:03:05,400 --> 00:03:08,040
Speaker 3: piece of hardware at the end of the project. I

48
00:03:08,080 --> 00:03:10,840
Speaker 3: think if you have that analytical brain, you like to

49
00:03:10,919 --> 00:03:14,560
Speaker 3: solve problems, you'd like to get things working. You can

50
00:03:14,600 --> 00:03:15,880
Speaker 3: do that in soet desertwork.

51
00:03:15,960 --> 00:03:19,320
Speaker 2: But as being someone coming from a hardware background mean

52
00:03:19,360 --> 00:03:21,959
Speaker 2: that you think about problems in a different way.

53
00:03:22,600 --> 00:03:26,239
Speaker 3: I think one thing that you do from a hardware background,

54
00:03:26,240 --> 00:03:29,359
Speaker 3: and especially a chip background, is a chip spin and

55
00:03:29,480 --> 00:03:33,360
Speaker 3: costs millions of dollars. So you're a lot more likely

56
00:03:33,480 --> 00:03:36,560
Speaker 3: to make sure everything has a great chance of being

57
00:03:36,600 --> 00:03:39,080
Speaker 3: perfect from the get go. Or if you start kind

58
00:03:39,080 --> 00:03:42,120
Speaker 3: of from a software background, your general mindset is I

59
00:03:42,120 --> 00:03:43,760
Speaker 3: don't know, try this, see if it works. I don't know.

60
00:03:43,800 --> 00:03:45,840
Speaker 3: Try that is, if it work, and you're kind of iterated,

61
00:03:45,960 --> 00:03:48,520
Speaker 3: or to iterate. Chip people are a little more uptight

62
00:03:48,600 --> 00:03:52,120
Speaker 3: about Okay, if this first round of the chip breaks

63
00:03:52,360 --> 00:03:55,320
Speaker 3: costs us from building another new round of the chip.

64
00:03:55,200 --> 00:03:58,680
Speaker 2: Yeah, so you're a little more You guys are spend more.

65
00:03:58,520 --> 00:04:04,880
Speaker 3: Time planning and planning verifying, tons of time verifying. Yeah.

66
00:04:05,320 --> 00:04:09,280
Speaker 2: So you began your career as you look backward, yes, correct,

67
00:04:09,320 --> 00:04:10,440
Speaker 2: And you were there for how many years?

68
00:04:10,520 --> 00:04:11,840
Speaker 3: I was there for thirty two years?

69
00:04:12,520 --> 00:04:15,680
Speaker 2: Yes, And your last job there was I was leading.

70
00:04:15,520 --> 00:04:18,200
Speaker 3: The software defining cloud business. I had grown up a

71
00:04:18,240 --> 00:04:23,720
Speaker 3: hardware guy. I had done all kinds of hardware projects,

72
00:04:23,760 --> 00:04:28,320
Speaker 3: big complicated Unix servers and things like that, and then came,

73
00:04:28,640 --> 00:04:30,440
Speaker 3: you know, grew out of R and D and more

74
00:04:30,440 --> 00:04:35,200
Speaker 3: into the business realm, and then I'm much an innovator

75
00:04:35,240 --> 00:04:39,680
Speaker 3: at heart. I really like innovating new concepts things like that.

76
00:04:39,960 --> 00:04:43,160
Speaker 3: And what I learned is I enjoyed innovating business models

77
00:04:43,200 --> 00:04:46,320
Speaker 3: and software projects as much as I did hardware products

78
00:04:47,120 --> 00:04:50,840
Speaker 3: and projects, and so getting teams inspired towards doing that

79
00:04:51,240 --> 00:04:53,599
Speaker 3: was really a deep fascination for me. So I ended

80
00:04:53,680 --> 00:04:57,400
Speaker 3: up doing a fantastic variety of experiences and had a

81
00:04:57,440 --> 00:05:01,640
Speaker 3: successful run and honestly retired in tending to retire and

82
00:05:02,000 --> 00:05:04,360
Speaker 3: do some of my outside activities and things like that.

83
00:05:04,880 --> 00:05:07,280
Speaker 2: And then how long did you stay retired before IBM

84
00:05:07,320 --> 00:05:07,760
Speaker 2: can close?

85
00:05:08,040 --> 00:05:12,480
Speaker 3: Almost two years? And when I first got at ALL,

86
00:05:12,720 --> 00:05:15,720
Speaker 3: I thought, no, I'm having too much fun. But I

87
00:05:15,760 --> 00:05:20,360
Speaker 3: would say three things really got me thinking hard about it.

88
00:05:20,440 --> 00:05:24,120
Speaker 3: One the industry that we're in, the IT industry. I

89
00:05:24,160 --> 00:05:26,200
Speaker 3: think it's the golden age. And what I mean by

90
00:05:26,240 --> 00:05:29,919
Speaker 3: that is for twenty years of that career, it is

91
00:05:30,000 --> 00:05:32,200
Speaker 3: kind of in the back office, say make sure that

92
00:05:32,240 --> 00:05:35,520
Speaker 3: stuff doesn't crash, and can you please reduce the cost

93
00:05:35,560 --> 00:05:37,599
Speaker 3: as much as possible, because it's not that important to

94
00:05:37,640 --> 00:05:41,080
Speaker 3: the main business. It's just a back office function. You

95
00:05:41,120 --> 00:05:43,400
Speaker 3: can see it right now. It is at the forefront

96
00:05:43,440 --> 00:05:47,719
Speaker 3: of all business revolution. It happened with the Internet. It

97
00:05:47,800 --> 00:05:51,480
Speaker 3: happened again with cloud and how that changed every ounce

98
00:05:51,520 --> 00:05:54,560
Speaker 3: of business, not just IT business, but all business. And

99
00:05:54,600 --> 00:05:57,719
Speaker 3: I think it's happening again with AI. So to be

100
00:05:57,839 --> 00:06:00,440
Speaker 3: in that career that long and to miss the kind

101
00:06:00,440 --> 00:06:02,920
Speaker 3: of this age where it's like this is front and center.

102
00:06:03,000 --> 00:06:06,680
Speaker 3: This changes everything about all businesses, not just technology businesses.

103
00:06:07,400 --> 00:06:11,200
Speaker 3: I was kind of feeling like, Gosh, you trained to

104
00:06:11,279 --> 00:06:16,320
Speaker 3: be in these really awesome environments, why wouldn't you do

105
00:06:16,400 --> 00:06:18,640
Speaker 3: that for a little while longer while you still can

106
00:06:18,720 --> 00:06:22,359
Speaker 3: do it. That combined with IBM and IBM seeing the

107
00:06:22,440 --> 00:06:26,640
Speaker 3: talent pool, the brilliant people at IBM, I worked with

108
00:06:26,680 --> 00:06:29,359
Speaker 3: a ton of brilliant people before I saw a chance

109
00:06:29,400 --> 00:06:32,000
Speaker 3: to work with even a larger staff of brilliant people.

110
00:06:32,120 --> 00:06:35,320
Speaker 3: And then the assets that IBM had, which is, you know,

111
00:06:35,320 --> 00:06:38,080
Speaker 3: they'd already been doing a lot of experimentation in AI,

112
00:06:38,200 --> 00:06:43,400
Speaker 3: they're working in quantum, the deep, rich heritage of successful projects.

113
00:06:43,400 --> 00:06:45,440
Speaker 3: I thought, who wouldn't want to kind of see if

114
00:06:45,440 --> 00:06:48,560
Speaker 3: they could be part of that next great wave of IBM.

115
00:06:48,680 --> 00:06:51,200
Speaker 3: And so I kind of decided, all right, I'm going

116
00:06:51,240 --> 00:06:53,279
Speaker 3: to put the outside interest on hold for a while

117
00:06:53,279 --> 00:06:54,320
Speaker 3: and get back in the game.

118
00:06:54,440 --> 00:06:57,760
Speaker 2: Along between the phone call, the first phone call and

119
00:06:57,800 --> 00:06:58,800
Speaker 2: you say, yes.

120
00:06:58,600 --> 00:07:03,520
Speaker 3: It was a while, was probably six months. Arvind's our CEO,

121
00:07:03,720 --> 00:07:05,839
Speaker 3: teases me about that a lot. Yeah, he was like,

122
00:07:05,920 --> 00:07:07,800
Speaker 3: I don't think six months. Is that long? It took

123
00:07:07,800 --> 00:07:11,680
Speaker 3: a while you're retirement, I know. Yeah, it's one thing

124
00:07:11,720 --> 00:07:14,360
Speaker 3: to compare. I'm working here and doing this stuff versus

125
00:07:14,360 --> 00:07:17,720
Speaker 3: working there. It's really hard to compare. I'm doing exactly

126
00:07:17,760 --> 00:07:19,720
Speaker 3: as I want to do every single day. When I

127
00:07:19,760 --> 00:07:21,440
Speaker 3: wake up and now I'm not going to get to

128
00:07:21,440 --> 00:07:24,240
Speaker 3: do that again. It took a while for me to

129
00:07:24,240 --> 00:07:27,040
Speaker 3: get over and I thought, I can't miss this wave,

130
00:07:27,080 --> 00:07:31,360
Speaker 3: and I'm really really happy that I did, because we're

131
00:07:31,400 --> 00:07:34,920
Speaker 3: doing some amazing fun things and I'm getting challenged in

132
00:07:34,960 --> 00:07:36,880
Speaker 3: ways that I never did, so it's really fun.

133
00:07:37,120 --> 00:07:40,160
Speaker 2: Talk a little bit about your job here at IBM.

134
00:07:40,360 --> 00:07:44,480
Speaker 2: You oversee a kind of massive portfolio.

135
00:07:44,640 --> 00:07:47,560
Speaker 3: It's a big group, so I run the Infrastructure organization.

136
00:07:47,600 --> 00:07:50,920
Speaker 3: There's three main groups of products at IBM. There's the

137
00:07:51,000 --> 00:07:54,720
Speaker 3: Infrastructure group, which I run, the Software group, and the

138
00:07:54,760 --> 00:07:59,880
Speaker 3: Consulting group. And infrastructure is built up of mainframes, which

139
00:07:59,880 --> 00:08:04,160
Speaker 3: is called our Z portfolio, our servers which is our

140
00:08:04,200 --> 00:08:08,520
Speaker 3: power portfolio storage. By the way, those businesses include the

141
00:08:08,560 --> 00:08:10,880
Speaker 3: supply chain to build all of that stuff, so that's

142
00:08:10,880 --> 00:08:14,680
Speaker 3: in the group. Then I have the worldwide Customer Support Organization.

143
00:08:14,760 --> 00:08:18,560
Speaker 3: It's called TLS Technology life Cycle Services, which is a

144
00:08:18,560 --> 00:08:21,960
Speaker 3: network of about thirteen thousand people around the globe that

145
00:08:22,040 --> 00:08:24,120
Speaker 3: make sure that everything runs and works when you buy

146
00:08:24,200 --> 00:08:27,800
Speaker 3: IBM products. And then also our IBM Cloud, which is

147
00:08:28,360 --> 00:08:33,080
Speaker 3: how we host applications and deliver as a service products

148
00:08:33,080 --> 00:08:35,839
Speaker 3: for our client base, so there's a lot. I think

149
00:08:35,840 --> 00:08:37,640
Speaker 3: it's about forty five thousand people total.

150
00:08:38,040 --> 00:08:42,600
Speaker 2: Do those components of the infrastructure group are they aligned

151
00:08:42,600 --> 00:08:46,079
Speaker 2: in their trajectory or do they are they on different paths?

152
00:08:46,120 --> 00:08:48,760
Speaker 2: And I'm just curious what so navigattle of both.

153
00:08:48,800 --> 00:08:51,400
Speaker 3: It's interesting you would ask that because I think of

154
00:08:51,679 --> 00:08:55,840
Speaker 3: all of the challenges coming to the new company, there

155
00:08:55,840 --> 00:08:58,560
Speaker 3: were things I expected, things that they didn't expect. But

156
00:08:59,640 --> 00:09:04,000
Speaker 3: getting that culture right in that group has been a

157
00:09:04,040 --> 00:09:10,120
Speaker 3: big challenge. IBM has a great culture toward quality products,

158
00:09:10,200 --> 00:09:14,200
Speaker 3: toward emphasizing passion for the client and making sure that

159
00:09:14,240 --> 00:09:17,840
Speaker 3: the client is happy, and for delivering innovation on a

160
00:09:17,880 --> 00:09:21,239
Speaker 3: scale that you know, for more than one hundred years

161
00:09:21,000 --> 00:09:26,079
Speaker 3: has been extremely powerful. But with success comes some challenges.

162
00:09:26,440 --> 00:09:28,560
Speaker 3: And with that success you can tend to get a

163
00:09:28,600 --> 00:09:32,160
Speaker 3: little bit insular, like you don't keep an eye on

164
00:09:32,200 --> 00:09:34,920
Speaker 3: the competition as well, you can get more siloed, where

165
00:09:35,280 --> 00:09:37,319
Speaker 3: you know, this is my business unit, this is my

166
00:09:37,480 --> 00:09:40,680
Speaker 3: business unit, I compete with the other business unit. That's

167
00:09:40,720 --> 00:09:44,280
Speaker 3: not a good thing when when you're a company and

168
00:09:44,320 --> 00:09:48,000
Speaker 3: you can get really risk averse, meaning you feel like, hey,

169
00:09:48,040 --> 00:09:50,080
Speaker 3: this is a successful business I don't want to do

170
00:09:50,120 --> 00:09:52,200
Speaker 3: anything to mess it up, so I don't need to

171
00:09:52,200 --> 00:09:55,079
Speaker 3: try new things. Well, that's exactly the recipe to kind

172
00:09:55,080 --> 00:09:57,360
Speaker 3: of be shrinking, and infrastructure had been shrinking for a

173
00:09:57,400 --> 00:10:01,120
Speaker 3: little while, and so a lot of what the challenge

174
00:10:01,200 --> 00:10:04,319
Speaker 3: was for me was to invigorate that risk taking and

175
00:10:05,480 --> 00:10:07,840
Speaker 3: get to a growth mindset where you're trying new things

176
00:10:07,880 --> 00:10:10,840
Speaker 3: and seeing what works and what doesn't work, and changing

177
00:10:10,840 --> 00:10:13,160
Speaker 3: some of the models, like investing a little bit less

178
00:10:13,160 --> 00:10:16,640
Speaker 3: in hardware for some software differentiation that goes into the hardware.

179
00:10:16,720 --> 00:10:20,199
Speaker 3: So it's been very successful so far, and it's been

180
00:10:20,200 --> 00:10:22,080
Speaker 3: a good journey. It's almost four years now.

181
00:10:22,440 --> 00:10:25,400
Speaker 2: Give me an example of what was a really hard

182
00:10:25,440 --> 00:10:27,160
Speaker 2: problem that you've dealt with in those four years.

183
00:10:27,160 --> 00:10:30,079
Speaker 3: So, boy, a really hard problem?

184
00:10:30,160 --> 00:10:33,120
Speaker 2: An interesting and are you interesting is a better word

185
00:10:33,160 --> 00:10:33,480
Speaker 2: than art.

186
00:10:33,760 --> 00:10:35,760
Speaker 3: One of the first things that I kind of chewed

187
00:10:35,800 --> 00:10:38,440
Speaker 3: on a little bit is I talked about how we

188
00:10:38,480 --> 00:10:41,600
Speaker 3: have Z power and storage. The Z and power product

189
00:10:41,640 --> 00:10:44,400
Speaker 3: lines are well known in the industry. Is is really

190
00:10:44,440 --> 00:10:47,760
Speaker 3: fit for purpose computing that have strengths that you know

191
00:10:47,880 --> 00:10:51,800
Speaker 3: Z runs you know most of the world's economic backbone,

192
00:10:51,840 --> 00:10:55,440
Speaker 3: and if you use a credit card, ninety percent of

193
00:10:55,480 --> 00:10:58,959
Speaker 3: credit card transactions for the globe. Go through these Z

194
00:10:59,440 --> 00:11:02,000
Speaker 3: mainframes there in every bank there. You know, it's a

195
00:11:02,000 --> 00:11:04,800
Speaker 3: big business. It's well known in the industry. Same with power,

196
00:11:05,240 --> 00:11:10,160
Speaker 3: very tuned and optimized for smaller operations than our giant

197
00:11:10,280 --> 00:11:16,240
Speaker 3: Z mainframes, but really mission critical workloads for retail, for insurance,

198
00:11:16,240 --> 00:11:19,440
Speaker 3: for banking, for all of that. Our storage business not

199
00:11:19,559 --> 00:11:22,360
Speaker 3: so well known. In fact, when I came I thought,

200
00:11:22,600 --> 00:11:25,000
Speaker 3: did they have storage? Well, I have storage when I

201
00:11:25,160 --> 00:11:27,600
Speaker 3: come into I. So I got online and I thought,

202
00:11:27,679 --> 00:11:29,480
Speaker 3: it's still hard for me to tell did they have

203
00:11:29,600 --> 00:11:31,920
Speaker 3: storage or not? Now I own a storage business. So

204
00:11:32,000 --> 00:11:33,560
Speaker 3: one of the things was not just to get the

205
00:11:33,600 --> 00:11:37,200
Speaker 3: market perception up, but to invest in that business. Because

206
00:11:37,240 --> 00:11:39,920
Speaker 3: if you look at infrastructure overall around the globe, it's

207
00:11:39,960 --> 00:11:43,520
Speaker 3: growing at five percent a year. The infrastructure business had

208
00:11:43,559 --> 00:11:46,240
Speaker 3: been kind of flat to declining, and so a challenge

209
00:11:46,280 --> 00:11:48,240
Speaker 3: was how do we grab onto the growth. Well, one

210
00:11:48,240 --> 00:11:50,960
Speaker 3: of the biggest growth areas due to the explosion of

211
00:11:51,040 --> 00:11:53,640
Speaker 3: data in the world is storage. So what do you

212
00:11:53,679 --> 00:11:55,880
Speaker 3: do to kind of get on that growth rate. So

213
00:11:55,880 --> 00:11:59,520
Speaker 3: we did a lot of reinvigoration of the innovation in

214
00:11:59,559 --> 00:12:02,439
Speaker 3: that a lot of software value, add a lot of

215
00:12:02,760 --> 00:12:05,839
Speaker 3: doubling down on the things that are working. Portfolio rationalization,

216
00:12:05,920 --> 00:12:08,160
Speaker 3: where you segment the market and you say, okay, we're

217
00:12:08,280 --> 00:12:10,080
Speaker 3: going to do less of this and really go big

218
00:12:10,120 --> 00:12:13,000
Speaker 3: in these areas. And that's been probably the most dramatic

219
00:12:13,000 --> 00:12:16,360
Speaker 3: turnaround inside the group. Is our storage thing. When you

220
00:12:16,400 --> 00:12:19,120
Speaker 3: say it's a hard problem, it's not just oh, you know,

221
00:12:19,160 --> 00:12:23,720
Speaker 3: how do we do the math? No, it's cultural. It's strategy,

222
00:12:23,800 --> 00:12:25,960
Speaker 3: and how do you get the strategy set. It's segmentation,

223
00:12:26,080 --> 00:12:29,199
Speaker 3: it's product strategy at a granular level across a bunch

224
00:12:29,240 --> 00:12:32,559
Speaker 3: of dimensions, and then putting the investment behind it. It's

225
00:12:32,600 --> 00:12:34,800
Speaker 3: a big challenge. It takes a long time, but it's working.

226
00:12:34,880 --> 00:12:36,839
Speaker 3: So we're happy with Yeah.

227
00:12:36,520 --> 00:12:38,480
Speaker 2: Tell me give me a little bit of perspective on

228
00:12:39,040 --> 00:12:42,840
Speaker 2: you've been there four years. Imagine we're having this conversation

229
00:12:42,920 --> 00:12:46,760
Speaker 2: four years ago. Yeah, what sorts of things have happened

230
00:12:46,800 --> 00:12:50,360
Speaker 2: over the last four years that have surprised you that

231
00:12:50,400 --> 00:12:53,360
Speaker 2: you didn't see come as we had exactly the same

232
00:12:53,360 --> 00:12:54,520
Speaker 2: conversation four years ago.

233
00:12:55,920 --> 00:12:57,640
Speaker 3: No, because I didn't know what was in I'll tell

234
00:12:57,640 --> 00:13:01,480
Speaker 3: you some of the biggest surprises I thought from the outside,

235
00:13:01,920 --> 00:13:05,119
Speaker 3: and you know, you hear from a lot of customers,

236
00:13:05,280 --> 00:13:08,360
Speaker 3: especially ten years ago. We're all going to cloud. We're

237
00:13:08,400 --> 00:13:10,760
Speaker 3: all so I thought, well, I wonder if the mainframe

238
00:13:10,840 --> 00:13:13,800
Speaker 3: business is struggling. When I get inside of there, I

239
00:13:13,840 --> 00:13:16,240
Speaker 3: found the opposite to be true. The mainframe business is

240
00:13:16,280 --> 00:13:20,400
Speaker 3: actually flourishing because transaction demand across the globe has done

241
00:13:20,480 --> 00:13:23,960
Speaker 3: nothing but grow. And even more surprising was the level

242
00:13:23,960 --> 00:13:27,000
Speaker 3: of innovation that the team was already doing in mainframes

243
00:13:27,440 --> 00:13:31,920
Speaker 3: before I got here was astounding. For example, we have AI.

244
00:13:32,800 --> 00:13:37,280
Speaker 3: They were building AI technology into the mainframe processors three

245
00:13:37,360 --> 00:13:40,640
Speaker 3: years before chat GPT made everybody talk about it in

246
00:13:40,679 --> 00:13:44,760
Speaker 3: the industry, So that was really pleasantly surprising. So that

247
00:13:44,920 --> 00:13:49,760
Speaker 3: was wonderful. Other surprises. I knew about the kind of

248
00:13:49,760 --> 00:13:52,880
Speaker 3: the IP of IBM and the mystique in that, and

249
00:13:52,960 --> 00:13:55,440
Speaker 3: I used to joke with people, especially on the outside,

250
00:13:55,440 --> 00:13:56,880
Speaker 3: I said, I can't wait to get in there and

251
00:13:56,880 --> 00:13:59,680
Speaker 3: see what's in the big blue toolbox? Right, what are

252
00:13:59,720 --> 00:14:03,040
Speaker 3: all the things they have going on? I way underestimated

253
00:14:03,240 --> 00:14:05,640
Speaker 3: the size of the big blue toolbox and what was

254
00:14:05,679 --> 00:14:10,280
Speaker 3: in there, meaning the amount of really hardcore research that

255
00:14:10,320 --> 00:14:13,839
Speaker 3: we're still doing into how to build chips and how

256
00:14:13,880 --> 00:14:16,960
Speaker 3: to get to things beyond two nanimeter and that kind

257
00:14:16,960 --> 00:14:22,960
Speaker 3: of capability packaging industry leading packaging technologies, and that's in

258
00:14:23,000 --> 00:14:25,760
Speaker 3: my hardware kind of patch quantum. The next thing that

259
00:14:25,840 --> 00:14:29,720
Speaker 3: will come after we're done talking about AI. You know,

260
00:14:30,440 --> 00:14:33,480
Speaker 3: all of those things were surprising, But it wasn't just that.

261
00:14:33,560 --> 00:14:36,120
Speaker 3: It was then the software innovations that are going on

262
00:14:36,280 --> 00:14:41,080
Speaker 3: heavy investment in AI technologies before it was really popular

263
00:14:41,120 --> 00:14:43,760
Speaker 3: to be talking about that. But as I saw that,

264
00:14:43,840 --> 00:14:46,480
Speaker 3: I thought this is going to be really fun. Because

265
00:14:46,520 --> 00:14:49,000
Speaker 3: I had a good feel for where the industry was going.

266
00:14:49,840 --> 00:14:52,360
Speaker 3: I just didn't and I knew, man, I know that

267
00:14:52,480 --> 00:14:55,120
Speaker 3: talent is really good and there's brilliant people there, but

268
00:14:55,160 --> 00:14:59,080
Speaker 3: I didn't know the level of IP frankly that IBM

269
00:14:59,200 --> 00:15:02,080
Speaker 3: had at its disposal, And now you're seeing that in

270
00:15:02,160 --> 00:15:05,359
Speaker 3: things like Watson X and things like AI in mainframes,

271
00:15:05,400 --> 00:15:05,800
Speaker 3: et cetera.

272
00:15:06,200 --> 00:15:09,160
Speaker 2: Building on that, since you brought up AI, can you

273
00:15:09,200 --> 00:15:13,080
Speaker 2: walk me through what has to happen from your perspective,

274
00:15:13,120 --> 00:15:19,200
Speaker 2: from the infrastructure perspective to make the AI explosion work? Yeah,

275
00:15:19,240 --> 00:15:21,360
Speaker 2: so everyone wants to do more of this stuff. Yes,

276
00:15:21,440 --> 00:15:23,920
Speaker 2: clearly there has to be some underpinning of it.

277
00:15:24,320 --> 00:15:28,440
Speaker 3: Yeah, I would tell you, you know, I think that people

278
00:15:28,600 --> 00:15:30,680
Speaker 3: feel like where we're at right now in the AI

279
00:15:30,760 --> 00:15:33,480
Speaker 3: journey had to do with one specific piece of software.

280
00:15:33,840 --> 00:15:37,960
Speaker 3: I think the inflection point for that whole thing really

281
00:15:38,880 --> 00:15:42,600
Speaker 3: at its root was around hardware, meaning the algorithms needed

282
00:15:42,640 --> 00:15:45,240
Speaker 3: to do larger language models. And all of that had

283
00:15:45,280 --> 00:15:47,920
Speaker 3: been around, they'd been talked about in the industry, but

284
00:15:47,960 --> 00:15:51,000
Speaker 3: at some point you hit a tipping point of hardware

285
00:15:51,040 --> 00:15:53,680
Speaker 3: capability where it's like, oh, now we can do this

286
00:15:53,800 --> 00:15:57,200
Speaker 3: in a broof force way, massive amounts of matrix math

287
00:15:57,280 --> 00:16:00,200
Speaker 3: to get weights correct so that you can do you know,

288
00:16:00,280 --> 00:16:03,600
Speaker 3: the right level of predictions that enable large language models.

289
00:16:03,960 --> 00:16:06,520
Speaker 3: And once we got to that horsepower. And that's why

290
00:16:06,560 --> 00:16:09,320
Speaker 3: you hear about giant GPUs that are driving this and

291
00:16:09,680 --> 00:16:11,920
Speaker 3: the sales of those, et cetera. It's because we just

292
00:16:12,000 --> 00:16:13,920
Speaker 3: barely got over the hump where you can do these

293
00:16:13,960 --> 00:16:19,240
Speaker 3: big hard things in terms of hardware capability to do it.

294
00:16:19,360 --> 00:16:22,200
Speaker 2: Give me a layman, give me a sense of when

295
00:16:22,240 --> 00:16:24,400
Speaker 2: you say there was a kind of threshold where suddenly

296
00:16:24,400 --> 00:16:25,560
Speaker 2: these things became possible.

297
00:16:25,680 --> 00:16:28,480
Speaker 3: Yeah, I don't know if there's an exact number, But

298
00:16:29,640 --> 00:16:31,720
Speaker 3: and more basic question that I get from a lot

299
00:16:31,760 --> 00:16:34,440
Speaker 3: of people. You know, my friends and family outside is

300
00:16:34,760 --> 00:16:39,480
Speaker 3: why GPUs. What does a GPU, a graphics processor have

301
00:16:39,600 --> 00:16:44,000
Speaker 3: to do with AI. It's not, Well, graphics processors are

302
00:16:44,040 --> 00:16:47,760
Speaker 3: really good at this thing matrix math, because they're figuring

303
00:16:47,840 --> 00:16:51,200
Speaker 3: out how do I map a pixel? And as I

304
00:16:51,320 --> 00:16:55,480
Speaker 3: move an object across the screen, it's essentially matrix math

305
00:16:55,560 --> 00:16:58,240
Speaker 3: to figure out, Okay, what does what does this pixel

306
00:16:58,320 --> 00:17:01,040
Speaker 3: on a screen look like? And what's it's doing? And

307
00:17:01,080 --> 00:17:04,560
Speaker 3: as you know, we've gotten more high resolution graphics, more

308
00:17:04,640 --> 00:17:07,680
Speaker 3: high resolution monitors, et cetera. It's a lot more pixels

309
00:17:07,720 --> 00:17:09,640
Speaker 3: and a lot more math and a lot more matrix

310
00:17:09,680 --> 00:17:12,480
Speaker 3: math about how you compute that. The first big thing

311
00:17:12,560 --> 00:17:15,200
Speaker 3: that kind of started to look like that, it turns out,

312
00:17:15,400 --> 00:17:18,880
Speaker 3: was crypto and crypto mining, and so you saw graphics

313
00:17:18,920 --> 00:17:22,119
Speaker 3: companies starting to sell to crypto. The technology got to

314
00:17:22,160 --> 00:17:24,480
Speaker 3: a certain point and there were use cases like bitcoin

315
00:17:24,560 --> 00:17:26,679
Speaker 3: and that that kind of said, hey, we need to

316
00:17:26,720 --> 00:17:29,000
Speaker 3: do a lot of this matrix math to be able

317
00:17:29,000 --> 00:17:32,040
Speaker 3: to do that. So graphic chips were a natural fit

318
00:17:32,119 --> 00:17:34,800
Speaker 3: and that kind of sustain But meanwhile, behind the scenes,

319
00:17:34,800 --> 00:17:39,240
Speaker 3: a lot of this AI AI is about numeric calculations

320
00:17:39,320 --> 00:17:43,000
Speaker 3: having to do with weights and matrices that say you know,

321
00:17:43,440 --> 00:17:46,879
Speaker 3: giant consolidated things that predict what's going to kind of

322
00:17:46,920 --> 00:17:49,320
Speaker 3: happen based on what other things have happened, just like

323
00:17:49,400 --> 00:17:53,920
Speaker 3: predicting where pixel goes. But it's really about being able

324
00:17:54,000 --> 00:17:56,800
Speaker 3: to do enough data in jest to be able to

325
00:17:56,840 --> 00:17:59,919
Speaker 3: do and then the calculations to be able to simple

326
00:18:00,080 --> 00:18:04,240
Speaker 3: five things like entire sets of language or giant chunks

327
00:18:04,280 --> 00:18:06,639
Speaker 3: of the Internet, to get enough waitings in there to

328
00:18:06,640 --> 00:18:09,320
Speaker 3: be able to say, Okay, we can predict what you

329
00:18:09,320 --> 00:18:12,720
Speaker 3: would say in this language based on all of the

330
00:18:12,800 --> 00:18:15,720
Speaker 3: volumes of stuff that we've seen that when you start

331
00:18:15,720 --> 00:18:18,960
Speaker 3: talking like this, the next word is likely, oh it's this. Yeah.

332
00:18:19,000 --> 00:18:21,240
Speaker 2: So, but my point is to get to that point.

333
00:18:21,359 --> 00:18:25,159
Speaker 2: That's threshold. We got there because was it because we

334
00:18:25,240 --> 00:18:28,080
Speaker 2: simply threw a lot more resources at the problem or

335
00:18:28,119 --> 00:18:32,399
Speaker 2: is it because the underlying technology got suddenly or gradually

336
00:18:32,560 --> 00:18:33,720
Speaker 2: so much more efficient.

337
00:18:33,840 --> 00:18:37,080
Speaker 3: It's always yes and yes. But you know, the industry

338
00:18:37,160 --> 00:18:39,320
Speaker 3: for a lot of years would talk about Moore's law.

339
00:18:39,800 --> 00:18:43,560
Speaker 2: Well, quick, will you define for us More's law for

340
00:18:43,600 --> 00:18:45,159
Speaker 2: those of those who's forgotten it.

341
00:18:45,400 --> 00:18:48,560
Speaker 3: Yeah, So Gordon Moore at Intel coined this thing. It

342
00:18:48,600 --> 00:18:53,320
Speaker 3: was basically that the horsepower I'm going to translate it

343
00:18:53,400 --> 00:18:59,040
Speaker 3: roughly of technology will double every couple of years. We're

344
00:18:59,080 --> 00:19:01,760
Speaker 3: still on Moore's law. More's law changed a little bit.

345
00:19:02,160 --> 00:19:05,000
Speaker 3: For a while, it was always about frequency. Things would

346
00:19:05,000 --> 00:19:08,159
Speaker 3: go faster, faster, faster. That kind of petered out. But

347
00:19:08,240 --> 00:19:11,440
Speaker 3: what happened is, rather than faster, faster, faster, we did

348
00:19:11,480 --> 00:19:14,760
Speaker 3: more and more and more. So rather than one operating

349
00:19:14,880 --> 00:19:18,720
Speaker 3: unit going a lot faster on its throughput, you put

350
00:19:18,760 --> 00:19:21,199
Speaker 3: ten operating units on a chip, now you put one

351
00:19:21,280 --> 00:19:24,320
Speaker 3: hundred operating units on a chip, now a thousand. Some

352
00:19:24,400 --> 00:19:29,040
Speaker 3: of these problems, the matrix math problems scaled parallel extremely well.

353
00:19:29,040 --> 00:19:31,320
Speaker 3: You don't have to do something really fast, you just

354
00:19:31,440 --> 00:19:33,440
Speaker 3: have to do a lot of the similar things in

355
00:19:33,480 --> 00:19:36,439
Speaker 3: parallel at the same time. So again that kind of

356
00:19:36,440 --> 00:19:39,159
Speaker 3: that extension of Moore's law, more and more hardware on

357
00:19:39,240 --> 00:19:40,640
Speaker 3: a chip to be able to do more and more

358
00:19:40,640 --> 00:19:43,879
Speaker 3: of those calculations in parallel and come up with it.

359
00:19:44,000 --> 00:19:47,120
Speaker 2: And we said, yeah, was that threshold predictable? In other words,

360
00:19:47,119 --> 00:19:50,000
Speaker 2: see people in the industry, like you sit down X

361
00:19:50,040 --> 00:19:52,080
Speaker 2: number of years ago and say, when we get here,

362
00:19:53,080 --> 00:19:54,840
Speaker 2: AI is going to become much more of a.

363
00:19:55,480 --> 00:20:02,080
Speaker 3: It's funny the horsepower that very pretty. The use cases

364
00:20:02,560 --> 00:20:05,160
Speaker 3: not always so easy to kind of figure out. That's

365
00:20:05,200 --> 00:20:08,840
Speaker 3: where the human spirit kind of gets involved. I think

366
00:20:08,880 --> 00:20:10,800
Speaker 3: for some people that say, oh, I saw that coming,

367
00:20:11,080 --> 00:20:14,199
Speaker 3: but people have been predicting kind of the rise of

368
00:20:14,280 --> 00:20:17,199
Speaker 3: AI for twenty five years. Oh well, then when we

369
00:20:17,280 --> 00:20:19,119
Speaker 3: get to this next gener oh when we get here,

370
00:20:19,320 --> 00:20:22,280
Speaker 3: it kind of hadn't happened. There's always a magic point

371
00:20:23,200 --> 00:20:25,520
Speaker 3: where you kind of get to where the technology and

372
00:20:25,600 --> 00:20:28,000
Speaker 3: the use case and somebody does something to kind of

373
00:20:28,040 --> 00:20:30,760
Speaker 3: make it catch on. And I think we're at one

374
00:20:30,760 --> 00:20:32,679
Speaker 3: of those moments in AI for sure right now. And

375
00:20:32,720 --> 00:20:34,919
Speaker 3: I don't think it's you know, people that have said, oh,

376
00:20:34,920 --> 00:20:38,760
Speaker 3: this is just the latest wave of you know, I've

377
00:20:38,800 --> 00:20:41,280
Speaker 3: heard this about a lot of technologies, but AI is

378
00:20:41,320 --> 00:20:43,960
Speaker 3: the technology the future, and it always will be. I

379
00:20:44,080 --> 00:20:46,760
Speaker 3: used to hear that. You're not hearing that now, right,

380
00:20:46,800 --> 00:20:50,480
Speaker 3: It's like, no, it's primetime. It will change everything, just

381
00:20:50,640 --> 00:20:52,680
Speaker 3: like some of these other things changed everything.

382
00:20:52,920 --> 00:20:57,359
Speaker 2: I noticed it if personally when I speak somewhere or

383
00:20:57,400 --> 00:21:01,000
Speaker 2: I'm listening in an audience somewhere. Over the last let's

384
00:21:01,000 --> 00:21:05,520
Speaker 2: say twelve months, there's always a whole bunch of AI

385
00:21:05,640 --> 00:21:08,520
Speaker 2: questions yes, And if I go back two years ago,

386
00:21:08,560 --> 00:21:09,760
Speaker 2: there were no AI questions.

387
00:21:09,840 --> 00:21:10,040
Speaker 3: Yes.

388
00:21:10,640 --> 00:21:12,960
Speaker 2: Now my question is, so there's been this explosion on

389
00:21:13,000 --> 00:21:17,399
Speaker 2: the in popular fascination with what's going on AI. It

390
00:21:17,480 --> 00:21:20,480
Speaker 2: seems like the last year. I agree with you in

391
00:21:20,560 --> 00:21:26,840
Speaker 2: your world, when did the explosion of conversation around this start.

392
00:21:27,560 --> 00:21:36,880
Speaker 3: It's I love this question because IBM had a fairly

393
00:21:37,040 --> 00:21:42,119
Speaker 3: big effort and business called Watson before Watson X. And

394
00:21:42,160 --> 00:21:45,080
Speaker 3: this is going back kind of ten years. I'll give

395
00:21:45,080 --> 00:21:47,920
Speaker 3: you another kind of example. I knew about a lot

396
00:21:47,920 --> 00:21:51,359
Speaker 3: of tablet technology before there was an iPad, a lot.

397
00:21:51,600 --> 00:21:53,960
Speaker 3: For ten years, there were a lot, but it kind

398
00:21:53,960 --> 00:21:57,640
Speaker 3: of takes a magic combination of the technology, the user experienced,

399
00:21:57,680 --> 00:22:00,520
Speaker 3: the software, and the need and the market for it

400
00:22:00,560 --> 00:22:02,719
Speaker 3: to kind of go. Now it's the thing. Now we

401
00:22:02,760 --> 00:22:05,359
Speaker 3: all have either an iPad or we have the Google

402
00:22:05,400 --> 00:22:08,320
Speaker 3: equivalent to And so I think this is a little

403
00:22:08,400 --> 00:22:11,720
Speaker 3: like that. Meaning IBM was on the right track with Watson.

404
00:22:12,240 --> 00:22:14,639
Speaker 3: Some of the hardware wasn't there, the use cases weren't

405
00:22:14,680 --> 00:22:17,280
Speaker 3: exactly figured out, some of the early use cases didn't

406
00:22:17,320 --> 00:22:20,399
Speaker 3: pan out perfectly. But the good news about that is

407
00:22:20,840 --> 00:22:23,680
Speaker 3: it's back to that culture of risk taking. You don't

408
00:22:23,720 --> 00:22:26,280
Speaker 3: look back on that and say, oh, we shouldn't have

409
00:22:26,320 --> 00:22:27,639
Speaker 3: done that, that was a bad idea. I know you

410
00:22:27,640 --> 00:22:29,360
Speaker 3: look back on that and say, what did we learn?

411
00:22:29,680 --> 00:22:31,879
Speaker 3: How should we try something new? How would we pivot

412
00:22:31,920 --> 00:22:34,159
Speaker 3: this time? That's what we've done with Watson X, and

413
00:22:35,359 --> 00:22:38,119
Speaker 3: now that's a growing, healthy piece of our business and

414
00:22:38,280 --> 00:22:40,760
Speaker 3: very important our strategic picture. So we're all in.

415
00:22:41,280 --> 00:22:47,800
Speaker 2: I've always investigated by the gap between insider sense of

416
00:22:47,840 --> 00:22:50,040
Speaker 2: what is happening in an outsider sense, like.

417
00:22:50,119 --> 00:22:53,080
Speaker 3: It absolutely is that in this case, we've all been

418
00:22:53,160 --> 00:22:57,160
Speaker 3: talking about and thinking about AI and is it time

419
00:22:57,200 --> 00:23:00,080
Speaker 3: for that and what does this mean? Et cetera. And

420
00:23:00,119 --> 00:23:02,760
Speaker 3: yet none of us really predicted that actual moment, which

421
00:23:02,840 --> 00:23:06,480
Speaker 3: is kind of you know, early twenty twenty two where

422
00:23:06,520 --> 00:23:10,480
Speaker 3: it was like, oh, now you have a simple human

423
00:23:10,560 --> 00:23:16,320
Speaker 3: interface of software innovation combined with large language models. There's

424
00:23:16,359 --> 00:23:19,560
Speaker 3: a moment there where you're like, oh, Unlike, you know,

425
00:23:19,560 --> 00:23:21,240
Speaker 3: I think all of us are frustrated if we ask

426
00:23:21,280 --> 00:23:23,680
Speaker 3: our phone, hey, tell me about this and it says

427
00:23:24,240 --> 00:23:26,199
Speaker 3: I found this on the web page. That does you

428
00:23:26,320 --> 00:23:28,000
Speaker 3: no good. But you know, all of a sudden, with

429
00:23:29,400 --> 00:23:31,800
Speaker 3: Chad GPT and some of these other things, you could

430
00:23:31,800 --> 00:23:33,879
Speaker 3: ask a question, it would give you a clear answer.

431
00:23:33,920 --> 00:23:36,560
Speaker 3: Sometimes is wrong, but at least it was like I'm

432
00:23:36,600 --> 00:23:38,439
Speaker 3: getting an answer rather than hey, I don't know if

433
00:23:38,440 --> 00:23:42,120
Speaker 3: there's some references. Good luck to you, and that's really changing.

434
00:23:42,680 --> 00:23:47,159
Speaker 3: Talk about the kind of macro trends that are going

435
00:23:47,240 --> 00:23:52,159
Speaker 3: to shape your infrastructure battle. Yeah, we've talked about if

436
00:23:52,160 --> 00:23:54,119
Speaker 3: you already, but I'm actually going to go a little

437
00:23:54,160 --> 00:23:58,560
Speaker 3: different direction. So macro trans first. And this one has

438
00:23:58,600 --> 00:24:02,520
Speaker 3: been before even even this AI conversation, that we've had

439
00:24:03,000 --> 00:24:10,080
Speaker 3: explosion of data. As humans, we don't think exponentially very well.

440
00:24:10,119 --> 00:24:14,040
Speaker 3: We really struggle with exponential thinking. We think linearly, Oh,

441
00:24:14,040 --> 00:24:16,080
Speaker 3: there'll be more, there'll be more, they'll be more, But

442
00:24:16,160 --> 00:24:18,200
Speaker 3: we don't think well when it's like no, there'll be more,

443
00:24:18,240 --> 00:24:19,840
Speaker 3: and they'll be ten times more, and then there'll be

444
00:24:19,880 --> 00:24:22,440
Speaker 3: ten times that more. That's what's going on with data

445
00:24:22,680 --> 00:24:25,320
Speaker 3: right now in our industry. It's one of the reasons

446
00:24:25,320 --> 00:24:27,680
Speaker 3: that that storage business is doing so well is they're

447
00:24:27,760 --> 00:24:31,560
Speaker 3: just more and more and more data. You know, you'd say, well,

448
00:24:31,600 --> 00:24:33,600
Speaker 3: how can there be more data? It's just life and

449
00:24:33,600 --> 00:24:36,880
Speaker 3: that thing. The things that we care about, video captured

450
00:24:36,960 --> 00:24:41,120
Speaker 3: video images, you know, the you I don't know from

451
00:24:41,160 --> 00:24:44,240
Speaker 3: my parents, you needed a drawer with all your family photos.

452
00:24:44,280 --> 00:24:45,920
Speaker 3: Now we need gigabytes and gigabytes.

453
00:24:46,000 --> 00:24:48,320
Speaker 2: You knew how many pictures my wife has taken off

454
00:24:48,320 --> 00:24:52,880
Speaker 2: our children, you would exactly exactly, So that's your case. Now.

455
00:24:52,960 --> 00:24:55,600
Speaker 3: Think of companies who used to just think about their

456
00:24:55,640 --> 00:24:59,399
Speaker 3: transaction data. What's the ledger say that now have video

457
00:24:59,640 --> 00:25:02,800
Speaker 3: assets of all of their campaigns and their marketing. They're

458
00:25:02,800 --> 00:25:05,520
Speaker 3: trying to figure out, you know, what campaigns are working

459
00:25:05,560 --> 00:25:08,280
Speaker 3: the best. So it's just an explosion of data and

460
00:25:08,320 --> 00:25:12,119
Speaker 3: that's not going to stop. Dealing with that, and more importantly,

461
00:25:12,200 --> 00:25:17,679
Speaker 3: getting value from that data is a massive trend in

462
00:25:17,720 --> 00:25:21,879
Speaker 3: the industry. Second trend AI, and this is the AI.

463
00:25:22,080 --> 00:25:24,119
Speaker 3: Not like we were just talking about about how it

464
00:25:24,200 --> 00:25:26,400
Speaker 3: changes how I search for things or how I learn

465
00:25:26,440 --> 00:25:30,840
Speaker 3: about things. But I would argue, dealing with that data,

466
00:25:30,880 --> 00:25:33,439
Speaker 3: how do I figure out what's in all those video streams?

467
00:25:33,480 --> 00:25:36,199
Speaker 3: How do I figure out, Okay, I want all of

468
00:25:36,240 --> 00:25:39,240
Speaker 3: the chunks of my corporate video that have to do

469
00:25:39,320 --> 00:25:43,359
Speaker 3: with client buying some specific product or something. That's a

470
00:25:44,000 --> 00:25:46,320
Speaker 3: different problem. It's not just okay, We'll look it up

471
00:25:46,320 --> 00:25:49,880
Speaker 3: in a spreadsheet and here's the math associated with that

472
00:25:49,880 --> 00:25:52,280
Speaker 3: that is a huge trend in the industry. You're seeing

473
00:25:52,280 --> 00:25:54,520
Speaker 3: it play out in this regard. It's a little different

474
00:25:54,600 --> 00:25:58,359
Speaker 3: bent on the AI fraud detection is the one that

475
00:25:58,400 --> 00:26:01,400
Speaker 3: we cite in our mainframe. It's a similar problem where

476
00:26:01,640 --> 00:26:04,320
Speaker 3: it was kind of a traditional AI problem. Look up

477
00:26:04,359 --> 00:26:08,480
Speaker 3: a rule. You know, if somebody does two small transactions,

478
00:26:08,520 --> 00:26:10,720
Speaker 3: then a massive one it might be fraud, right because

479
00:26:10,720 --> 00:26:14,200
Speaker 3: they were seeing whether it were now to detect fraud,

480
00:26:14,240 --> 00:26:18,240
Speaker 3: you might be saying, okay to transactions then a huge one.

481
00:26:18,560 --> 00:26:22,399
Speaker 3: Plus does this entity have a real address? Second, is

482
00:26:22,440 --> 00:26:25,760
Speaker 3: there any web traffic on you know, better Business bureau

483
00:26:25,840 --> 00:26:27,640
Speaker 3: kind of things that says this is a bad business

484
00:26:27,720 --> 00:26:29,800
Speaker 3: that can help you with fraud. So it's a lot

485
00:26:29,880 --> 00:26:33,200
Speaker 3: more of a it's an expotent problem. It's a holistic

486
00:26:33,280 --> 00:26:36,000
Speaker 3: problem that it takes a lot more than just you know,

487
00:26:36,240 --> 00:26:39,080
Speaker 3: little chunks of rules, et cetera. And then the third

488
00:26:39,080 --> 00:26:42,920
Speaker 3: one you know, after AI, is the nature of hybrid

489
00:26:42,960 --> 00:26:46,600
Speaker 3: it or hybrid computing. For a while ten years ago

490
00:26:46,680 --> 00:26:49,879
Speaker 3: when cloud was on the rise, I think the notion

491
00:26:49,960 --> 00:26:53,159
Speaker 3: of hybrid computing basically having to do with things in

492
00:26:53,200 --> 00:26:57,199
Speaker 3: the cloud versus things that people still have on the

493
00:26:57,240 --> 00:27:01,840
Speaker 3: premises inside of business. It was almost religious argument. Now

494
00:27:01,960 --> 00:27:05,000
Speaker 3: it's no, it's the reality. And the reason is because

495
00:27:05,040 --> 00:27:08,359
Speaker 3: that data that I talked about is the lifeblood of

496
00:27:08,400 --> 00:27:13,439
Speaker 3: these companies, particularly IBM's companies are clients that usually that

497
00:27:13,560 --> 00:27:15,920
Speaker 3: data has to be secure, they have to be able

498
00:27:15,960 --> 00:27:18,840
Speaker 3: to get value from it. It is the lifeblood of

499
00:27:18,840 --> 00:27:20,600
Speaker 3: the company. If you go to an ATM and you

500
00:27:20,600 --> 00:27:24,080
Speaker 3: can't get your money out, you know, to our financial transactions,

501
00:27:24,800 --> 00:27:27,160
Speaker 3: if that lasts a day, you're probably going to change

502
00:27:27,160 --> 00:27:30,160
Speaker 3: banks immediately. So it's like life or death for these companies.

503
00:27:32,040 --> 00:27:36,520
Speaker 3: So having that hybrid infrastructure so that they can still

504
00:27:36,560 --> 00:27:40,000
Speaker 3: hold their data, you still interact with clouds and still

505
00:27:40,000 --> 00:27:42,960
Speaker 3: get value from it from AI. That's kind of the

506
00:27:43,080 --> 00:27:47,680
Speaker 3: magic where we play and it's a huge business opportunity.

507
00:27:47,960 --> 00:27:50,360
Speaker 3: It is a true inflection point for the industry.

508
00:27:51,440 --> 00:27:54,840
Speaker 2: I'm going to go back. I interrupted you when you

509
00:27:54,880 --> 00:27:57,000
Speaker 2: were in the middle of a rellion. We were talking

510
00:27:57,000 --> 00:28:01,679
Speaker 2: about what has to happen for for AI to scale

511
00:28:01,960 --> 00:28:04,879
Speaker 2: from the infrastructure standpoint. You gave one example that I

512
00:28:05,119 --> 00:28:07,480
Speaker 2: got you off on a tangent. Can you go back

513
00:28:07,520 --> 00:28:10,960
Speaker 2: and talk very so practically, like so I'm you know,

514
00:28:11,040 --> 00:28:14,480
Speaker 2: I'm a big company. I have all these dreams of AI,

515
00:28:15,000 --> 00:28:17,280
Speaker 2: of how I'm going to use this dratically. So give

516
00:28:17,320 --> 00:28:20,600
Speaker 2: me a very granular sense of the works you have

517
00:28:20,680 --> 00:28:23,160
Speaker 2: to do, yeah to make that dream possible.

518
00:28:23,600 --> 00:28:26,840
Speaker 3: So let me first say what the company has to do,

519
00:28:26,880 --> 00:28:28,840
Speaker 3: and then maybe I'll say, then how do I help them?

520
00:28:28,920 --> 00:28:31,080
Speaker 3: If that makes sense? So if I'm a company and

521
00:28:31,080 --> 00:28:33,080
Speaker 3: I want to do that, So it turns out I

522
00:28:33,119 --> 00:28:36,919
Speaker 3: am a company, meaning I want to use AI in

523
00:28:36,960 --> 00:28:40,800
Speaker 3: my processes. I mentioned that I have a global network

524
00:28:40,840 --> 00:28:45,000
Speaker 3: of thirteen thousand employees that support our infrastructure around the world.

525
00:28:45,400 --> 00:28:50,800
Speaker 3: That challenge is a great challenge for AI. That means

526
00:28:50,840 --> 00:28:55,400
Speaker 3: I have data for every customer situation for thirteen thousand

527
00:28:55,440 --> 00:28:58,880
Speaker 3: employees globally around the world on what was their problem,

528
00:28:58,920 --> 00:29:02,400
Speaker 3: how did we fix it, what next steps did they

529
00:29:02,400 --> 00:29:04,840
Speaker 3: have to do, how did they remediate that? That data

530
00:29:04,920 --> 00:29:07,400
Speaker 3: is extremely valuable to me because if I can get

531
00:29:07,440 --> 00:29:10,000
Speaker 3: better at doing that than anybody else in the world,

532
00:29:10,360 --> 00:29:12,680
Speaker 3: that brings my cost down. I sell more products, I

533
00:29:12,720 --> 00:29:15,600
Speaker 3: sell more service, I sell more anything. So what I

534
00:29:15,720 --> 00:29:17,640
Speaker 3: have to do to get there is I have to

535
00:29:17,640 --> 00:29:20,920
Speaker 3: figure out, Okay, what's my objective. I have a couple objectives.

536
00:29:20,960 --> 00:29:23,400
Speaker 3: One I want customers to be able to support themselves

537
00:29:23,400 --> 00:29:26,440
Speaker 3: without even calling me, first off, and I don't want

538
00:29:26,440 --> 00:29:29,960
Speaker 3: when they call for the first answer to come back

539
00:29:29,960 --> 00:29:32,520
Speaker 3: to be did you try rebooting? Because I think that

540
00:29:32,760 --> 00:29:35,560
Speaker 3: irritates every single one of us. Did you try? Of

541
00:29:35,600 --> 00:29:38,640
Speaker 3: course I tried rebooting. I've had a lap up, of course.

542
00:29:38,680 --> 00:29:42,840
Speaker 3: I well, okay, well then tell me, okay, what firmware version,

543
00:29:42,880 --> 00:29:44,960
Speaker 3: all that other stuff. Okay, we know this interaction. So

544
00:29:46,240 --> 00:29:48,160
Speaker 3: that's kind of the problem set. Do I want that

545
00:29:48,320 --> 00:29:51,600
Speaker 3: to be customers solving their own problems? Well, even for

546
00:29:51,760 --> 00:29:54,200
Speaker 3: my support agents, I want something in their pocket on

547
00:29:54,240 --> 00:29:57,160
Speaker 3: their phone where they say I'm seeing these symptoms. It says, oh,

548
00:29:57,400 --> 00:29:59,920
Speaker 3: this happening around the globe. Here's here's kind of specific.

549
00:30:00,160 --> 00:30:03,840
Speaker 3: So there's my problems. What does it mean for infrastructure

550
00:30:03,840 --> 00:30:07,720
Speaker 3: on the back end? So first I got to get

551
00:30:07,720 --> 00:30:10,480
Speaker 3: all that data together, right, all of those customer law,

552
00:30:10,520 --> 00:30:13,600
Speaker 3: all that customer support around the globe, et cetera. That

553
00:30:13,680 --> 00:30:16,040
Speaker 3: needs to be stored. That's a big set of data.

554
00:30:16,400 --> 00:30:19,560
Speaker 3: And some of it's not just fix and that kind

555
00:30:19,640 --> 00:30:22,080
Speaker 3: of thing. Some of it is, Okay, you know, what

556
00:30:22,200 --> 00:30:24,480
Speaker 3: was the firmware version? Who was the tech? Because it

557
00:30:24,520 --> 00:30:27,520
Speaker 3: can matter. Is this their first time fixing this problem?

558
00:30:27,520 --> 00:30:29,520
Speaker 3: Is that they're one hundred and fiftieth time. What's their level?

559
00:30:29,800 --> 00:30:34,520
Speaker 3: It's a very complicated problem. Ingesting all that data takes

560
00:30:34,680 --> 00:30:38,080
Speaker 3: an architecture. We have a product called Scale, which is

561
00:30:38,120 --> 00:30:41,200
Speaker 3: one of our storage projects that actually makes it easy

562
00:30:41,240 --> 00:30:44,400
Speaker 3: to ingest all that data, get it organized, et cetera,

563
00:30:44,840 --> 00:30:48,840
Speaker 3: and then have a model. It's a whole different process

564
00:30:48,920 --> 00:30:50,640
Speaker 3: to kind of say did we train our model? We

565
00:30:50,680 --> 00:30:52,840
Speaker 3: can train our own models inside of IBM. We have

566
00:30:52,880 --> 00:30:56,080
Speaker 3: a granite set of models. Those models we fine tune,

567
00:30:56,440 --> 00:30:58,840
Speaker 3: and then we inference based on those models. So we

568
00:30:58,880 --> 00:31:01,400
Speaker 3: can do that inferencing in our cloud I have a

569
00:31:01,440 --> 00:31:04,280
Speaker 3: cloud set of infrastructure or in my power servers, we

570
00:31:04,320 --> 00:31:09,160
Speaker 3: can do inferencing with our capabilities and say, okay, based

571
00:31:09,200 --> 00:31:12,160
Speaker 3: on what I'm saying, here's what the remediation you should

572
00:31:12,200 --> 00:31:15,360
Speaker 3: do for that customer. We already are doing that today.

573
00:31:15,400 --> 00:31:20,920
Speaker 3: We've seen over a third of our support calls have

574
00:31:21,040 --> 00:31:24,240
Speaker 3: had significant reduction in the amount of time that it

575
00:31:24,280 --> 00:31:27,680
Speaker 3: takes to resolve that support call. Just by what I

576
00:31:27,720 --> 00:31:28,720
Speaker 3: said right there.

577
00:31:28,840 --> 00:31:32,400
Speaker 2: That I've really been curious about this. If I had

578
00:31:32,400 --> 00:31:36,640
Speaker 2: reduced something like AI into that equation as you just did. Yeah,

579
00:31:36,680 --> 00:31:39,680
Speaker 2: and you said we've already seen a thirty percent, Say

580
00:31:39,680 --> 00:31:41,040
Speaker 2: did you say thirty percent reduction?

581
00:31:41,400 --> 00:31:45,960
Speaker 3: Thirty percent of our interactions have seen significant reduction in

582
00:31:46,360 --> 00:31:46,960
Speaker 3: those times?

583
00:31:47,080 --> 00:31:50,080
Speaker 2: Was that your primary goal to reduce the time of

584
00:31:50,120 --> 00:31:52,880
Speaker 2: the interaction? But it was if you if everything else

585
00:31:52,960 --> 00:31:55,240
Speaker 2: was the same all, but what you were doing was

586
00:31:55,280 --> 00:31:56,960
Speaker 2: shrinking the amount of time That would you.

587
00:31:56,920 --> 00:32:01,200
Speaker 3: Want one of the primary goals? So to us in

588
00:32:01,240 --> 00:32:04,880
Speaker 3: that business net promoter score kind of the satisfaction of

589
00:32:04,920 --> 00:32:08,360
Speaker 3: a client is the supreme goal. What makes them satisfied,

590
00:32:08,800 --> 00:32:11,800
Speaker 3: doesn't cost me a fortune, happens really quickly, and if

591
00:32:11,840 --> 00:32:14,920
Speaker 3: I can do it myself, I'd be thrilled. It affects

592
00:32:15,000 --> 00:32:17,360
Speaker 3: all of those right. It kind of says it got

593
00:32:17,400 --> 00:32:19,720
Speaker 3: resolved faster, it didn't cost me an arm and a leg,

594
00:32:19,760 --> 00:32:22,840
Speaker 3: because the deck was barely here, because it's a common problem,

595
00:32:23,160 --> 00:32:27,000
Speaker 3: or I solved it myself without even calling, So all

596
00:32:27,040 --> 00:32:29,160
Speaker 3: of those objectives would kind of hit across all so

597
00:32:29,200 --> 00:32:31,440
Speaker 3: that now you see it. So that's a little microcosm.

598
00:32:31,480 --> 00:32:33,880
Speaker 3: That's just me and my customer support business. Now think

599
00:32:33,960 --> 00:32:37,480
Speaker 3: of how many problems for businesses around the world there

600
00:32:37,480 --> 00:32:40,200
Speaker 3: are like that it's not a it's not like a

601
00:32:40,280 --> 00:32:44,640
Speaker 3: new AI application that changes the entire user experience. That's

602
00:32:45,520 --> 00:32:48,880
Speaker 3: those will come, But right now it's kind of practical,

603
00:32:48,960 --> 00:32:51,000
Speaker 3: which is, I just want to do what I'm doing

604
00:32:51,080 --> 00:32:54,840
Speaker 3: better and faster, and I can get immediate economic return

605
00:32:54,920 --> 00:32:55,600
Speaker 3: from those things.

606
00:32:55,600 --> 00:32:58,680
Speaker 2: How long how long did it take you to just

607
00:32:58,760 --> 00:33:02,120
Speaker 2: stick with that example of the customer reaction reducing thirty

608
00:33:02,120 --> 00:33:04,800
Speaker 2: percent of the time? How long from the very beginning

609
00:33:04,800 --> 00:33:08,160
Speaker 2: of that project yeap to that thirty percent reduction was?

610
00:33:08,160 --> 00:33:08,520
Speaker 2: How long?

611
00:33:09,160 --> 00:33:14,560
Speaker 3: Less than a year? And yeah, So one of the challenges,

612
00:33:14,840 --> 00:33:18,479
Speaker 3: and this is interesting with a very large organization, as

613
00:33:18,560 --> 00:33:21,000
Speaker 3: you can imagine, just like you're seeing in the industry,

614
00:33:21,880 --> 00:33:25,440
Speaker 3: we don't have a problem of generating ideas for how

615
00:33:25,520 --> 00:33:29,160
Speaker 3: AI could help us. We actually have a problem filtering

616
00:33:29,560 --> 00:33:33,400
Speaker 3: the thousands of ideas from our employees and from from everywhere.

617
00:33:33,400 --> 00:33:35,760
Speaker 3: It's like, hey, we could use AI to and filtering

618
00:33:35,800 --> 00:33:37,960
Speaker 3: down and saying, okay, which of these will have a

619
00:33:38,040 --> 00:33:41,800
Speaker 3: return on investment quickly and at a level that sustains

620
00:33:41,840 --> 00:33:44,560
Speaker 3: that's worth kind of going and investing in the infrastructure

621
00:33:44,560 --> 00:33:48,040
Speaker 3: and the software and kind of making that happen.

622
00:33:48,160 --> 00:33:51,440
Speaker 2: Is that unusual? If I talked to you twenty five

623
00:33:51,560 --> 00:33:53,880
Speaker 2: years ago and said, do you have a problem of

624
00:33:53,920 --> 00:33:55,280
Speaker 2: too many good ideas or too few?

625
00:33:55,320 --> 00:34:01,480
Speaker 3: What was you said in this specific carriot, Probably too few,

626
00:34:01,640 --> 00:34:05,040
Speaker 3: because at some point you reach diminishing returns. So, for example,

627
00:34:05,080 --> 00:34:09,360
Speaker 3: let's use this same example. Can those thirteen thousand technicians

628
00:34:09,400 --> 00:34:13,799
Speaker 3: go faster? Can they spend less time driving to the side.

629
00:34:13,800 --> 00:34:15,520
Speaker 3: I mean, there's only so much you can kind of

630
00:34:15,560 --> 00:34:17,920
Speaker 3: do on those things. But if you can get them

631
00:34:17,920 --> 00:34:20,200
Speaker 3: an answer to the problem and maybe even avoid them

632
00:34:20,239 --> 00:34:22,799
Speaker 3: having to visit at all because the client helped themselves,

633
00:34:23,160 --> 00:34:26,160
Speaker 3: that's a step function. So that's why people are kind

634
00:34:26,200 --> 00:34:30,600
Speaker 3: of talking about there's a business revolution coming with AI

635
00:34:30,719 --> 00:34:33,719
Speaker 3: where there are some step function changes that can be there.

636
00:34:33,760 --> 00:34:37,080
Speaker 3: And notice I didn't say I'm going to have less

637
00:34:37,120 --> 00:34:40,719
Speaker 3: of those agents. That's not my objective. My objective and

638
00:34:40,719 --> 00:34:43,120
Speaker 3: I think that's the fear in the industry about AI

639
00:34:43,200 --> 00:34:45,560
Speaker 3: is going to eliminate all the jobs. No, I just

640
00:34:45,640 --> 00:34:49,600
Speaker 3: created thirteen thousand superpowered agents that can do more right.

641
00:34:49,680 --> 00:34:51,880
Speaker 3: And so I'm not just going to support IBM products.

642
00:34:52,120 --> 00:34:54,280
Speaker 3: I'm going to go out and support other people's products

643
00:34:54,320 --> 00:34:55,960
Speaker 3: because I know how to do that really well. And

644
00:34:56,000 --> 00:34:58,759
Speaker 3: once I have the data on how to fix their problems.

645
00:34:59,200 --> 00:35:02,800
Speaker 3: I may just have a customer support business that's independent

646
00:35:02,840 --> 00:35:05,600
Speaker 3: of mind boxes. So you know, I think that's where

647
00:35:05,640 --> 00:35:08,160
Speaker 3: people sometimes get it wrong. And the AI thing is

648
00:35:08,760 --> 00:35:12,719
Speaker 3: it's like, you know, do word processing eliminate the need

649
00:35:12,800 --> 00:35:18,200
Speaker 3: for writers? No? It enabled writing instead of mucking around

650
00:35:18,200 --> 00:35:20,840
Speaker 3: with mimeographic machines and click and click typewriters.

651
00:35:20,920 --> 00:35:23,400
Speaker 2: It may have enabled too much writing. Yeah, maybe maybe

652
00:35:24,040 --> 00:35:27,200
Speaker 2: can I give you a hypothetical? Uh? And I asked

653
00:35:27,239 --> 00:35:29,279
Speaker 2: this because I read I was at some convers and

654
00:35:29,280 --> 00:35:31,360
Speaker 2: I ran into some guy from the I R S

655
00:35:32,520 --> 00:35:36,000
Speaker 2: who was really, really, really really excited about AI. So

656
00:35:36,080 --> 00:35:40,239
Speaker 2: let's suppose they call you up and they say, you're

657
00:35:40,280 --> 00:35:43,920
Speaker 2: going to talk to the I R S. Okay, I

658
00:35:44,000 --> 00:35:48,879
Speaker 2: call you up and I say, Rick, Uh, clearly there's

659
00:35:48,920 --> 00:35:52,480
Speaker 2: something that we could do for the I R S

660
00:35:52,520 --> 00:35:53,360
Speaker 2: if we work together.

661
00:35:53,560 --> 00:35:53,759
Speaker 3: Yeah.

662
00:35:53,760 --> 00:35:54,640
Speaker 2: Who would your answer me?

663
00:35:55,920 --> 00:35:56,400
Speaker 3: Of course?

664
00:35:57,080 --> 00:35:57,160
Speaker 2: No.

665
00:35:57,280 --> 00:36:01,000
Speaker 3: I think we sell to a lot of government agencies.

666
00:36:01,360 --> 00:36:05,120
Speaker 3: Can imagine in the business that we're in, we enable

667
00:36:05,280 --> 00:36:08,600
Speaker 3: a lot of social security transactions and things like that

668
00:36:08,600 --> 00:36:12,799
Speaker 3: through our mainframes. And I think, you know, we're in

669
00:36:12,840 --> 00:36:16,319
Speaker 3: the business of helping whatever client get the most out

670
00:36:16,320 --> 00:36:18,320
Speaker 3: of their data and be able to secure it and

671
00:36:18,880 --> 00:36:21,760
Speaker 3: and be able to do analytics with this. And IRS

672
00:36:21,800 --> 00:36:23,959
Speaker 3: has a heck of a lot of data, so yes,

673
00:36:24,000 --> 00:36:24,759
Speaker 3: we would help them.

674
00:36:25,120 --> 00:36:26,840
Speaker 2: Do you know how the amount of data they have

675
00:36:26,960 --> 00:36:29,000
Speaker 2: compares to some of the corporate clients you've.

676
00:36:29,080 --> 00:36:32,040
Speaker 3: I don't know specifically for the IRS how much data

677
00:36:32,120 --> 00:36:34,040
Speaker 3: they have, but I would assume it's a whole lot.

678
00:36:34,360 --> 00:36:37,719
Speaker 3: It's mountains. But but that's our business. I mean, it's

679
00:36:37,760 --> 00:36:41,279
Speaker 3: interesting sometimes people have that what's the most you know,

680
00:36:41,320 --> 00:36:45,279
Speaker 3: what what is it that that IBM has that's of

681
00:36:45,480 --> 00:36:49,279
Speaker 3: great value? Is it a server? Is it a storage array?

682
00:36:49,400 --> 00:36:51,920
Speaker 3: Is it you know, software and all that. What we

683
00:36:52,040 --> 00:36:56,440
Speaker 3: have is the most important entities in the world have

684
00:36:56,600 --> 00:37:00,520
Speaker 3: their data on our stuff. The most important data in

685
00:37:00,560 --> 00:37:03,759
Speaker 3: the world. It's not you know, pictures of your grandkids

686
00:37:03,840 --> 00:37:06,120
Speaker 3: and things like that. Generally for us, it's all of

687
00:37:06,160 --> 00:37:09,520
Speaker 3: the financial transactions that happen globally, right, It's all of

688
00:37:09,560 --> 00:37:12,560
Speaker 3: the it's the world's economy is kind of running through

689
00:37:13,160 --> 00:37:16,319
Speaker 3: our systems, and so we take that really seriously. You know,

690
00:37:17,200 --> 00:37:19,440
Speaker 3: you would be distraught if you lost one photo on

691
00:37:19,480 --> 00:37:22,279
Speaker 3: your laptop or whatever. But you know, if we lose

692
00:37:22,320 --> 00:37:25,359
Speaker 3: a transaction, like somebody moves a big amount of money

693
00:37:25,400 --> 00:37:28,120
Speaker 3: and it's like, well, don't know what happened there. It

694
00:37:28,200 --> 00:37:31,520
Speaker 3: is a massive deal, right, so that doesn't happen.

695
00:37:31,640 --> 00:37:33,160
Speaker 2: But I want to go back to my irs example

696
00:37:33,200 --> 00:37:37,080
Speaker 2: for US, Yes, so one, is it reasonable to assume

697
00:37:37,360 --> 00:37:42,399
Speaker 2: that you could that somebody IBM or somebody else could

698
00:37:42,400 --> 00:37:44,880
Speaker 2: in a short period of time put together not just

699
00:37:44,960 --> 00:37:50,520
Speaker 2: the AI capability to audit returns, but also this the

700
00:37:50,560 --> 00:37:53,799
Speaker 2: infrastructure support for that in a reasonable amount of time

701
00:37:53,800 --> 00:37:56,520
Speaker 2: for a reasonable amount of cost. Or is it overall?

702
00:37:56,640 --> 00:37:59,480
Speaker 2: Is it going to the moon? Or is it it?

703
00:37:59,680 --> 00:38:03,560
Speaker 3: Definitely? I mean, so we're already doing that kind of

704
00:38:03,600 --> 00:38:08,880
Speaker 3: thing right across a network of banks and others, essentially

705
00:38:09,000 --> 00:38:12,920
Speaker 3: all credit card transactions for all of the world to

706
00:38:12,960 --> 00:38:16,200
Speaker 3: go through our systems, so that in some ways is

707
00:38:16,239 --> 00:38:19,720
Speaker 3: more volume than the datch returns of the US people.

708
00:38:19,840 --> 00:38:23,040
Speaker 3: And they're W two's and all that stuff, and we

709
00:38:23,120 --> 00:38:25,920
Speaker 3: do that stuff too. I try not to describe it

710
00:38:25,960 --> 00:38:28,160
Speaker 3: too much in detail, but we definitely do a lot

711
00:38:28,200 --> 00:38:33,840
Speaker 3: of that. In fact, I think most of if you think, okay,

712
00:38:33,840 --> 00:38:37,000
Speaker 3: what is super critical data, who would be doing the

713
00:38:37,040 --> 00:38:41,080
Speaker 3: business transaction processing? It is most likely US in almost

714
00:38:41,160 --> 00:38:45,520
Speaker 3: all cases, whether it's government things or private or banks

715
00:38:45,840 --> 00:38:47,960
Speaker 3: or that kind of thing. That's what we do.

716
00:38:48,280 --> 00:38:50,280
Speaker 2: Rick we're going to end with the where we always

717
00:38:50,400 --> 00:38:52,240
Speaker 2: end with a couple of quick fire questions.

718
00:38:52,320 --> 00:38:53,319
Speaker 3: Okay, here we go.

719
00:38:54,200 --> 00:38:57,280
Speaker 2: What single piece of advice would you give to businesses

720
00:38:57,320 --> 00:38:59,879
Speaker 2: trying to use AI in an effective way?

721
00:39:00,120 --> 00:39:03,680
Speaker 3: The simple version is get started. By get started, I

722
00:39:03,719 --> 00:39:07,760
Speaker 3: mean think of what is something that I want to improve.

723
00:39:07,880 --> 00:39:10,040
Speaker 3: The things that we have traction on right now in

724
00:39:10,080 --> 00:39:16,000
Speaker 3: the market are around business process, automation, digital labor, those

725
00:39:16,120 --> 00:39:19,320
Speaker 3: kind of things. But my other little piece of advice

726
00:39:19,360 --> 00:39:21,360
Speaker 3: there is keep it simple to begin with. You're going

727
00:39:21,440 --> 00:39:24,120
Speaker 3: to learn a lot, but getting started means you'll start

728
00:39:24,160 --> 00:39:27,960
Speaker 3: that learning curve. I even advise my friends like, Hey,

729
00:39:27,960 --> 00:39:30,520
Speaker 3: should I be playing around with some of this AI stuff?

730
00:39:30,560 --> 00:39:33,000
Speaker 3: And I say yeah, because I think it will help

731
00:39:33,040 --> 00:39:35,719
Speaker 3: you start to be more comfortable and you may find

732
00:39:35,760 --> 00:39:37,759
Speaker 3: a use case personally for that. I think the same

733
00:39:37,880 --> 00:39:40,920
Speaker 3: is true for businesses. The first step in that journey

734
00:39:40,960 --> 00:39:44,680
Speaker 3: is always with what data. Notice when I talked about

735
00:39:44,680 --> 00:39:49,000
Speaker 3: our customer support people, I thought about, Okay, what's the data.

736
00:39:49,160 --> 00:39:51,719
Speaker 3: The data is all of those logs of all of

737
00:39:51,719 --> 00:39:54,920
Speaker 3: those service engagements around the world, and what could I

738
00:39:54,960 --> 00:39:56,759
Speaker 3: do with that? Well, I could use that to get

739
00:39:56,800 --> 00:40:00,719
Speaker 3: to a knowledge base that really helps hopefully that I

740
00:40:00,760 --> 00:40:03,319
Speaker 3: can do it in multiple languages because it's global and

741
00:40:03,400 --> 00:40:05,920
Speaker 3: I can you know, all of those things. That was

742
00:40:06,000 --> 00:40:08,760
Speaker 3: kind of my data sent That one's not super simple,

743
00:40:08,800 --> 00:40:11,560
Speaker 3: but we've had a lot of experience in AI for

744
00:40:11,640 --> 00:40:14,520
Speaker 3: other people that might just be how do I automate

745
00:40:14,640 --> 00:40:18,600
Speaker 3: filling out travel expense reports for my company? We can

746
00:40:18,600 --> 00:40:21,040
Speaker 3: help people that we have consulting, we have wats and

747
00:40:21,200 --> 00:40:23,319
Speaker 3: X tools. We can do that like this, and we're

748
00:40:23,360 --> 00:40:26,479
Speaker 3: doing it globally for people around the world. Pick that thing.

749
00:40:26,560 --> 00:40:29,520
Speaker 3: What's the data you have? In that case, it's data

750
00:40:29,520 --> 00:40:31,839
Speaker 3: of expense reports and it's like, okay, we can help

751
00:40:31,840 --> 00:40:34,120
Speaker 3: you automate that for people where they could do it

752
00:40:34,239 --> 00:40:38,040
Speaker 3: just by you know, a verbal interface. What did you spend,

753
00:40:38,120 --> 00:40:40,120
Speaker 3: where did you go, who you were you with? Okay,

754
00:40:40,200 --> 00:40:42,480
Speaker 3: we filled out your travel expense report for you and

755
00:40:42,520 --> 00:40:43,799
Speaker 3: you don't have to mess around with it.

756
00:40:43,960 --> 00:40:46,600
Speaker 2: So we were playing with this idea where we would

757
00:40:46,920 --> 00:40:50,279
Speaker 2: pick a business and go in there and do it

758
00:40:50,320 --> 00:40:51,360
Speaker 2: would be AI makeover.

759
00:40:51,640 --> 00:40:52,640
Speaker 3: Yeah, I love that.

760
00:40:52,960 --> 00:40:55,960
Speaker 2: What's okay, what's the what is the ideal business to do.

761
00:40:56,680 --> 00:40:58,200
Speaker 2: We only have a couple months. We don't want to

762
00:40:58,200 --> 00:41:00,480
Speaker 2: spend a kajillion dollars. We want to be able to

763
00:41:00,520 --> 00:41:04,239
Speaker 2: show tangibly and quickly what AI can do. What's that

764
00:41:04,360 --> 00:41:06,120
Speaker 2: ideal business to do? That in it can be a

765
00:41:06,120 --> 00:41:09,120
Speaker 2: small business, but we're not talking. This isn't a grand corporate.

766
00:41:08,800 --> 00:41:13,560
Speaker 3: Thing there, ah boy, small business that we could do

767
00:41:13,640 --> 00:41:17,840
Speaker 3: and hey, I make over. Customer support is one of

768
00:41:17,840 --> 00:41:21,239
Speaker 3: my favorites because it's it's it's I have it on

769
00:41:21,280 --> 00:41:24,600
Speaker 3: the business side where I provide customer support. I have

770
00:41:24,680 --> 00:41:27,240
Speaker 3: it on the consumer side, where it drives me nuts

771
00:41:27,560 --> 00:41:30,680
Speaker 3: when I have to go through thirty layers of phone menus.

772
00:41:31,200 --> 00:41:33,239
Speaker 3: Speak to an agent, speak to an agent, speak to

773
00:41:33,280 --> 00:41:37,480
Speaker 3: an agent. That for any business, I think is just

774
00:41:37,719 --> 00:41:39,960
Speaker 3: ripe to be able to kind of say, why do

775
00:41:40,040 --> 00:41:42,319
Speaker 3: I have to click through these manuscent messages? I just

776
00:41:42,360 --> 00:41:44,960
Speaker 3: need to tell you in human language, here's the issue,

777
00:41:44,960 --> 00:41:47,720
Speaker 3: and I'll be really good about telling you details about

778
00:41:48,239 --> 00:41:50,680
Speaker 3: you know, I tried to set up this thing for

779
00:41:50,840 --> 00:41:52,640
Speaker 3: my bank and I do da da da da da.

780
00:41:52,880 --> 00:41:56,560
Speaker 3: They can go through all the menus automate that process.

781
00:41:56,840 --> 00:41:59,160
Speaker 3: I think it would change everything because all that frustration

782
00:41:59,360 --> 00:42:02,719
Speaker 3: is a consumer would go down dramatically and it's all,

783
00:42:03,239 --> 00:42:06,240
Speaker 3: you know, why are you making me the beep booth,

784
00:42:06,440 --> 00:42:10,560
Speaker 3: press one, offload press exactly, Well, don't offload to me,

785
00:42:10,800 --> 00:42:13,239
Speaker 3: offload to AI. We can help you with that.

786
00:42:13,520 --> 00:42:16,800
Speaker 2: Here's my version of that drives me crazy. Every morning

787
00:42:16,960 --> 00:42:20,000
Speaker 2: I go to the same coffee shop and I get

788
00:42:20,600 --> 00:42:22,719
Speaker 2: a cup of tea and a croissant.

789
00:42:23,080 --> 00:42:24,000
Speaker 3: And here's what happens.

790
00:42:24,000 --> 00:42:26,719
Speaker 2: The person has their screen and they go I go,

791
00:42:27,000 --> 00:42:35,239
Speaker 2: cup of tea, croissant, sparkling water, like at least twenty keystrokes,

792
00:42:36,000 --> 00:42:38,640
Speaker 2: and then like then the screen is turned around. Like

793
00:42:38,719 --> 00:42:41,080
Speaker 2: at this point we're like forty five seconds in, I'm like,

794
00:42:41,360 --> 00:42:43,279
Speaker 2: why is this? First of all, it's not for me.

795
00:42:43,360 --> 00:42:46,520
Speaker 2: All those keystrokes, it's there in turn right right, So

796
00:42:46,719 --> 00:42:48,320
Speaker 2: they're burdening me in order to service.

797
00:42:48,360 --> 00:42:50,080
Speaker 3: To back it, you should be able to walk in,

798
00:42:50,320 --> 00:42:52,319
Speaker 3: go up and they go, I'm olc them the same

799
00:42:52,360 --> 00:42:54,879
Speaker 3: thing and you just go yes, and then they boom,

800
00:42:55,040 --> 00:42:55,439
Speaker 3: We're done.

801
00:42:55,480 --> 00:42:57,600
Speaker 2: Can we do AI makeover of my coffee shop?

802
00:42:59,440 --> 00:43:03,160
Speaker 3: You notice I quickly jumped more to banks than your

803
00:43:03,239 --> 00:43:06,200
Speaker 3: coffee shop because I think I'm a business person, but

804
00:43:06,520 --> 00:43:08,680
Speaker 3: I'm not trying to kind of do a deal on

805
00:43:08,680 --> 00:43:09,520
Speaker 3: one coffee shop.

806
00:43:09,520 --> 00:43:11,719
Speaker 2: No, But this is interesting because it takes me back

807
00:43:11,760 --> 00:43:14,280
Speaker 2: to something you said that I thought was really important.

808
00:43:14,640 --> 00:43:17,359
Speaker 2: When you were talking about when you were using AI

809
00:43:17,520 --> 00:43:20,800
Speaker 2: and your customer service thing, it was clear that your goal,

810
00:43:20,960 --> 00:43:23,759
Speaker 2: you could have any number of goals going in. It

811
00:43:23,800 --> 00:43:27,280
Speaker 2: could be to cut costs, it could be to dramatically

812
00:43:27,320 --> 00:43:28,440
Speaker 2: improved profits.

813
00:43:29,000 --> 00:43:29,840
Speaker 3: Your goal, quite.

814
00:43:29,680 --> 00:43:32,560
Speaker 2: Specifically, was to improve the experience of your customer, right,

815
00:43:32,600 --> 00:43:33,920
Speaker 2: So you were using it to that.

816
00:43:34,120 --> 00:43:36,879
Speaker 3: All the other things come from that come from. That

817
00:43:37,120 --> 00:43:40,720
Speaker 3: is actually one of the beautiful pillars of the IBM

818
00:43:40,760 --> 00:43:44,360
Speaker 3: culture is delighting clients is actually where all of the

819
00:43:44,440 --> 00:43:45,560
Speaker 3: good stuff comes from.

820
00:43:45,600 --> 00:43:49,640
Speaker 2: So my coffee shop thing is the same principle. Right now,

821
00:43:49,880 --> 00:43:52,920
Speaker 2: they're making my customer experience worse and they don't want to.

822
00:43:54,400 --> 00:43:56,279
Speaker 3: Their eyes are glued to the special a.

823
00:43:56,239 --> 00:43:58,359
Speaker 2: Moment when I walk in and I want to say, Hi,

824
00:43:58,480 --> 00:44:03,520
Speaker 2: how are you doing conversation? You're too busy, busy, So like,

825
00:44:04,040 --> 00:44:05,839
Speaker 2: this is the same thing. If they had it, oh,

826
00:44:05,840 --> 00:44:08,279
Speaker 2: we this is if they understood they had an opportunity

827
00:44:08,320 --> 00:44:11,040
Speaker 2: to improve the experience of their customer experience.

828
00:44:11,080 --> 00:44:14,960
Speaker 3: I would not be surprised if a chain comes along

829
00:44:15,400 --> 00:44:17,880
Speaker 3: where that is their value proposition, I would not be

830
00:44:17,920 --> 00:44:22,520
Speaker 3: surprised at all. Yeah, right, So I mean and and

831
00:44:22,560 --> 00:44:26,280
Speaker 3: when those things kind of catch hold, it becomes a revolution.

832
00:44:26,520 --> 00:44:28,480
Speaker 2: You know, when the guy comes to do like to

833
00:44:28,480 --> 00:44:31,160
Speaker 2: redo your roof and they put a sign out front,

834
00:44:31,239 --> 00:44:33,840
Speaker 2: like you know, Joe's roofing. You guys could do the

835
00:44:33,840 --> 00:44:37,840
Speaker 2: same with my coffee shop. But like I'd be was

836
00:44:37,920 --> 00:44:45,759
Speaker 2: here exactly exactly, in five years, the main frame will

837
00:44:45,760 --> 00:44:48,640
Speaker 2: be dot dot going strong.

838
00:44:49,320 --> 00:44:54,840
Speaker 3: H the main frame going strong and with new capabilities,

839
00:44:54,920 --> 00:44:59,480
Speaker 3: continuous new capabilities. I think when we announced the last

840
00:44:59,560 --> 00:45:04,120
Speaker 3: versions six, the latest version, I should say, and we said, hey,

841
00:45:04,120 --> 00:45:07,880
Speaker 3: there's AI processing built into it. This was before everybody

842
00:45:07,920 --> 00:45:10,200
Speaker 3: was talking about that. I think a lot of people thought,

843
00:45:10,480 --> 00:45:13,360
Speaker 3: what's that for? And we did it specifically for traditional

844
00:45:13,360 --> 00:45:17,359
Speaker 3: AI fraud detection, et cetera. This next version, not only

845
00:45:17,400 --> 00:45:19,560
Speaker 3: do we have the traditional AI built in, but we

846
00:45:19,640 --> 00:45:22,799
Speaker 3: have optional cards that you can plug into it to

847
00:45:22,840 --> 00:45:26,640
Speaker 3: allow you to do large language models for the enhanced

848
00:45:26,680 --> 00:45:30,440
Speaker 3: fraud detection cases that we talked about, where you know,

849
00:45:30,520 --> 00:45:34,439
Speaker 3: it's more than just what transactions were happening. So if

850
00:45:34,440 --> 00:45:38,360
Speaker 3: you take that and say, okay, the next generations. We

851
00:45:38,440 --> 00:45:42,000
Speaker 3: have more transaction volume than we've ever had in mainframes. Today,

852
00:45:42,200 --> 00:45:46,000
Speaker 3: the business is growing, it's strong, we keep innovating. In

853
00:45:46,040 --> 00:45:47,600
Speaker 3: five years it'll be going strong.

854
00:45:47,719 --> 00:45:50,319
Speaker 2: But we're people. You're saying this in the context of

855
00:45:51,200 --> 00:45:53,399
Speaker 2: for years people were predicting, weren't they that the main

856
00:45:53,440 --> 00:45:54,480
Speaker 2: brand was going to go away.

857
00:45:56,360 --> 00:45:58,799
Speaker 3: There were pundits in the market that said everything will

858
00:45:58,800 --> 00:46:00,800
Speaker 3: go away there no one will ever have a box,

859
00:46:00,800 --> 00:46:03,480
Speaker 3: It'll all be online. I think this is something I've

860
00:46:03,560 --> 00:46:08,080
Speaker 3: learned big time in my long career. You know in

861
00:46:08,120 --> 00:46:12,359
Speaker 3: the IT industry is don't believe everything you hear. So

862
00:46:12,440 --> 00:46:16,520
Speaker 3: I went back for my master's degree at Stanford after

863
00:46:16,560 --> 00:46:21,160
Speaker 3: I had worked a while in as a hardware designer,

864
00:46:21,440 --> 00:46:24,280
Speaker 3: and everybody told me be sure to do your masters

865
00:46:24,280 --> 00:46:27,320
Speaker 3: in software. Hardware is dead. I went on to work

866
00:46:27,560 --> 00:46:31,120
Speaker 3: for thirty plus years in hardware and infrastructure. Now software

867
00:46:31,120 --> 00:46:33,560
Speaker 3: became important, and I'm glad I had that extra training

868
00:46:33,560 --> 00:46:36,360
Speaker 3: in software because it helped me in hardware. But hardware

869
00:46:36,440 --> 00:46:40,000
Speaker 3: wasn't dead. Then I heard all infrastructure will go into

870
00:46:40,040 --> 00:46:43,080
Speaker 3: the cloud. There won't be that hasn't happened. It's not happening.

871
00:46:43,360 --> 00:46:45,800
Speaker 3: Then I heard there will only be one cloud because

872
00:46:45,840 --> 00:46:48,280
Speaker 3: one of the players will dominate. There's not one cloud.

873
00:46:48,360 --> 00:46:52,600
Speaker 3: So I think it's as humans we like to oversimplify

874
00:46:52,640 --> 00:46:54,920
Speaker 3: and go, oh, it's all going to be this, and

875
00:46:55,000 --> 00:46:59,080
Speaker 3: kind of what I've learned is fit for purpose matters

876
00:46:59,120 --> 00:47:04,880
Speaker 3: in everything. It matters in size of infrastructure, it matters

877
00:47:04,880 --> 00:47:07,600
Speaker 3: in the stack that goes along with solving a specific

878
00:47:07,760 --> 00:47:10,960
Speaker 3: use case. If you're willing to design something that's the

879
00:47:10,960 --> 00:47:13,280
Speaker 3: best at that use case, If you're willing to design

880
00:47:13,360 --> 00:47:15,680
Speaker 3: the coffee shop that is the best at greeting me,

881
00:47:16,040 --> 00:47:17,920
Speaker 3: there's a spot for you, and there may be a

882
00:47:17,960 --> 00:47:21,840
Speaker 3: big business in doing that. So oversimplifying is really.

883
00:47:21,840 --> 00:47:25,040
Speaker 2: When you heard all those predictions, did you believe them

884
00:47:25,040 --> 00:47:25,799
Speaker 2: at the time.

885
00:47:27,200 --> 00:47:29,960
Speaker 3: They looked like they were trending in that direction. I'll

886
00:47:30,000 --> 00:47:32,680
Speaker 3: tell you some right now which might be useful. There

887
00:47:32,680 --> 00:47:35,080
Speaker 3: will only be one GPU company and they're going to

888
00:47:35,800 --> 00:47:38,400
Speaker 3: end up taking over the world. It's a pretty obvious answer.

889
00:47:38,440 --> 00:47:41,839
Speaker 3: Whose economic values risen dramatically. I don't think that's going

890
00:47:41,880 --> 00:47:44,320
Speaker 3: to be the case. In fact, I think that ninety

891
00:47:44,360 --> 00:47:50,000
Speaker 3: percent of processing for AI actually happen happens at inferencing,

892
00:47:50,440 --> 00:47:53,920
Speaker 3: and inferencing is not as GPU and hardware intensive as

893
00:47:53,960 --> 00:47:56,400
Speaker 3: the other things, and is a lot more amenable to

894
00:47:56,520 --> 00:47:59,600
Speaker 3: fit for purpose. So the model size will matter. The

895
00:47:59,680 --> 00:48:02,040
Speaker 3: tune matters a lot. As we're learning. We have a

896
00:48:02,040 --> 00:48:06,120
Speaker 3: product around instruct lab that's really focused on tuning. So

897
00:48:06,560 --> 00:48:08,520
Speaker 3: that was one thing is there'll be one GPU. The

898
00:48:08,560 --> 00:48:12,480
Speaker 3: other thing is that the biggest model will win. I

899
00:48:12,520 --> 00:48:14,879
Speaker 3: think is another thing that's kind of people are saying

900
00:48:14,960 --> 00:48:17,359
Speaker 3: right now. Don't believe that I believe they'll be fit

901
00:48:17,440 --> 00:48:20,520
Speaker 3: for purpose models. It takes a lot of money to

902
00:48:20,600 --> 00:48:24,120
Speaker 3: run to create a huge model, and then to run

903
00:48:24,160 --> 00:48:26,719
Speaker 3: a huge model, or to even infer off of a

904
00:48:26,840 --> 00:48:30,439
Speaker 3: huge model. I don't need a massive training GPU set

905
00:48:30,480 --> 00:48:34,160
Speaker 3: thing to solve my thirteen thousand people customer support issues.

906
00:48:34,200 --> 00:48:36,319
Speaker 3: So why would I feel like I got to go

907
00:48:36,440 --> 00:48:38,839
Speaker 3: farm that out for a big expensive thing. I can

908
00:48:38,880 --> 00:48:40,840
Speaker 3: do that on a small box. In some cases I

909
00:48:40,920 --> 00:48:42,640
Speaker 3: might even be able to do that on a laptop.

910
00:48:43,120 --> 00:48:45,000
Speaker 3: The other thing I'll say in this we are so

911
00:48:45,239 --> 00:48:47,960
Speaker 3: early innings in AI, A lot of things are going

912
00:48:48,000 --> 00:48:50,399
Speaker 3: to change. So anybody kind of saying it will all

913
00:48:50,440 --> 00:48:52,920
Speaker 3: be X, Y or Z, I just think you have

914
00:48:53,000 --> 00:48:54,920
Speaker 3: no idea how this is going to play out. And

915
00:48:55,600 --> 00:48:57,279
Speaker 3: it's up to us to go figure out how it

916
00:48:57,280 --> 00:48:57,799
Speaker 3: plays out.

917
00:48:58,080 --> 00:49:02,000
Speaker 2: Yeah, yeah, all right, in five years, AI will be

918
00:49:02,320 --> 00:49:03,160
Speaker 2: dot dot dot.

919
00:49:03,800 --> 00:49:10,160
Speaker 3: Still new. It will have moved a bunch in five years,

920
00:49:10,880 --> 00:49:15,160
Speaker 3: but the potential for the disruption in the world will

921
00:49:15,200 --> 00:49:18,320
Speaker 3: still will still be very early innings in that process.

922
00:49:18,440 --> 00:49:20,799
Speaker 3: And I think that's super important to realize. That's why

923
00:49:20,840 --> 00:49:24,359
Speaker 3: I say get started, start thinking about how that could change,

924
00:49:24,360 --> 00:49:27,200
Speaker 3: because it'll be some little things first, but it will

925
00:49:27,239 --> 00:49:28,320
Speaker 3: continue to snowball.

926
00:49:28,520 --> 00:49:34,000
Speaker 2: This is a common observation that we the invention of

927
00:49:34,040 --> 00:49:40,439
Speaker 2: the capability uh massively predates the understanding of the capability, right,

928
00:49:40,520 --> 00:49:44,560
Speaker 2: Like I love that. Yeah, Like, yes, recorded recording shows

929
00:49:44,680 --> 00:49:52,239
Speaker 2: on television is invented in the sixties. Probably we don't

930
00:49:52,280 --> 00:49:57,440
Speaker 2: really understand what it's used for until the oughts was

931
00:49:57,200 --> 00:49:59,600
Speaker 2: what's really good for is being able to tell a

932
00:49:59,640 --> 00:50:02,920
Speaker 2: story sequentially, Yes, over time, because you know that the

933
00:50:02,920 --> 00:50:04,719
Speaker 2: person will all have see in the episode before, so

934
00:50:04,760 --> 00:50:08,319
Speaker 2: you got the Sopranos And yes, yes, Hollywood wanted to

935
00:50:08,480 --> 00:50:11,680
Speaker 2: ban the VCR in the beginning. Yeah, because they thought

936
00:50:11,680 --> 00:50:13,680
Speaker 2: it was good. They thought the point of it was

937
00:50:14,000 --> 00:50:17,040
Speaker 2: thought the din understand No, no, no, it's storytelling. It's actually

938
00:50:17,120 --> 00:50:19,840
Speaker 2: your business is getting better. Yes, Yes, took them twenty

939
00:50:19,880 --> 00:50:21,719
Speaker 2: years to figure that out, which is to your point,

940
00:50:22,239 --> 00:50:24,400
Speaker 2: why would we know what AI was four and five yearso?

941
00:50:24,400 --> 00:50:26,520
Speaker 3: Well, that's why you hear people kind of say, oh

942
00:50:26,600 --> 00:50:29,279
Speaker 3: my gosh, AI, that's that will just eliminate jobs. No,

943
00:50:29,360 --> 00:50:31,440
Speaker 3: it'll make jobs better. That's how I view it.

944
00:50:31,560 --> 00:50:35,040
Speaker 2: Yeah, what's the number one thing that people misunderstand about AI?

945
00:50:35,200 --> 00:50:38,440
Speaker 3: Is that it that it'll I think that's that. That

946
00:50:38,480 --> 00:50:41,400
Speaker 3: would be the human kind of understanding part of it.

947
00:50:41,480 --> 00:50:44,799
Speaker 3: The technology part of it, I think would be what

948
00:50:44,960 --> 00:50:49,160
Speaker 3: I was talking about fit for purpose, meaning that it

949
00:50:49,200 --> 00:50:52,000
Speaker 3: isn't just going to be a GPU arms race all

950
00:50:52,040 --> 00:50:54,600
Speaker 3: of AI. I don't believe that at all. It will

951
00:50:54,680 --> 00:50:57,000
Speaker 3: change everything, but it's not just going to be a GPU.

952
00:50:56,800 --> 00:51:00,440
Speaker 2: Armed Next question, what advice would you give yourself ten

953
00:51:00,480 --> 00:51:03,440
Speaker 2: years ago to better prepare you for today? I'm changing

954
00:51:03,440 --> 00:51:08,040
Speaker 2: this question, Okay. I want to say, let's imagine that

955
00:51:09,400 --> 00:51:10,440
Speaker 2: what was your what.

956
00:51:10,400 --> 00:51:13,879
Speaker 3: College you to go to? I went to three of them.

957
00:51:14,080 --> 00:51:17,440
Speaker 3: My undergrad was Utah State University, my MBA was Santa

958
00:51:17,440 --> 00:51:20,600
Speaker 3: Clara University, and my master's in w was Stanford.

959
00:51:20,760 --> 00:51:24,000
Speaker 2: Okay, any one of those three culture up and says

960
00:51:24,360 --> 00:51:29,200
Speaker 2: we want you to give the commencement address and imagine

961
00:51:29,239 --> 00:51:31,919
Speaker 2: that it's it's it's let's just say, for the sake

962
00:51:31,920 --> 00:51:33,880
Speaker 2: of argument, it's just to the stamp people.

963
00:51:34,320 --> 00:51:35,640
Speaker 3: Those are the relevant parties here.

964
00:51:36,640 --> 00:51:37,839
Speaker 2: What do you tell them?

965
00:51:38,239 --> 00:51:44,520
Speaker 3: Boy? What do I tell them? Let's see. I think

966
00:51:44,600 --> 00:51:49,320
Speaker 3: I would start with life is a marathon, not a sprint.

967
00:51:49,719 --> 00:51:53,560
Speaker 3: It would be the first one. The second thing I

968
00:51:53,560 --> 00:51:56,840
Speaker 3: would say that in that spirit is be sure to

969
00:51:56,960 --> 00:52:02,120
Speaker 3: set yourself some big, hairy, audacious goals and don't be

970
00:52:02,320 --> 00:52:07,680
Speaker 3: overly disappointed if you don't hit them all. Going after

971
00:52:07,719 --> 00:52:10,400
Speaker 3: those big, hairy, audacious goals will get you on a

972
00:52:10,440 --> 00:52:14,640
Speaker 3: path where you will learn so much. You will achieve

973
00:52:14,719 --> 00:52:17,400
Speaker 3: more than you ever could imagine you would have achieved.

974
00:52:17,640 --> 00:52:19,680
Speaker 3: That's what the advice I give to my kids is,

975
00:52:20,080 --> 00:52:22,960
Speaker 3: set some big goals, get after it. You may or

976
00:52:22,960 --> 00:52:24,680
Speaker 3: may not achieve them, but you'll be better for the

977
00:52:24,680 --> 00:52:25,799
Speaker 3: whole process when you're done.

978
00:52:25,800 --> 00:52:28,200
Speaker 2: By the way, as someone whose kids are younger than yours,

979
00:52:29,000 --> 00:52:31,279
Speaker 2: is it actually useful to give you give advice to

980
00:52:31,320 --> 00:52:34,160
Speaker 2: your kids the points exercise TVD.

981
00:52:34,239 --> 00:52:36,200
Speaker 3: We're still on the journey, and I think we will

982
00:52:36,200 --> 00:52:39,880
Speaker 3: be for a long time. I don't know how.

983
00:52:39,719 --> 00:52:41,719
Speaker 2: Are you already using AI in your day to day

984
00:52:41,719 --> 00:52:42,240
Speaker 2: life today?

985
00:52:44,280 --> 00:52:48,440
Speaker 3: Personally, I would say it's replacing a good chunk of

986
00:52:48,520 --> 00:52:51,520
Speaker 3: my search. You know, I'm less likely to go blindly

987
00:52:51,840 --> 00:52:54,720
Speaker 3: stumbling through a bunch of web pages looking for stuff.

988
00:52:55,080 --> 00:52:57,440
Speaker 3: I'm more likely to ask a question from a few

989
00:52:57,600 --> 00:53:00,400
Speaker 3: AI engines kind of see get me in the right direction,

990
00:53:00,480 --> 00:53:03,120
Speaker 3: then I'll go bumble through a few things. At work,

991
00:53:03,560 --> 00:53:08,920
Speaker 3: I can tell you code development right now, we are

992
00:53:08,960 --> 00:53:13,000
Speaker 3: seeing massive improvements in code development and support. Products we

993
00:53:13,120 --> 00:53:17,920
Speaker 3: have like Watson Code Assistant that is really showing immediate

994
00:53:17,960 --> 00:53:21,160
Speaker 3: return for a code developers, and I think that will

995
00:53:21,200 --> 00:53:25,120
Speaker 3: again be a tool that increases productivity for code developers

996
00:53:25,480 --> 00:53:28,040
Speaker 3: immediately across the globe. Yeah.

997
00:53:28,160 --> 00:53:31,520
Speaker 2: Last question, what's the one skill that every technology leader

998
00:53:31,680 --> 00:53:33,879
Speaker 2: needs that has nothing to do with technology.

999
00:53:34,840 --> 00:53:38,960
Speaker 3: Being able to inspire a set of people toward a

1000
00:53:39,000 --> 00:53:42,520
Speaker 3: common goal and collaborate to achieve it. That's at the

1001
00:53:42,520 --> 00:53:46,359
Speaker 3: core of everything everything. That's a lovely way to end.

1002
00:53:47,000 --> 00:53:48,319
Speaker 3: Thank you so much, Rick, Thank you.

1003
00:53:50,840 --> 00:53:55,160
Speaker 2: This conversation left me excited. I'm now imagining the potential

1004
00:53:55,200 --> 00:53:57,960
Speaker 2: for new use cases for AI in all sorts of

1005
00:53:58,040 --> 00:54:01,680
Speaker 2: different businesses. Rick didn't seem sold on my idea of

1006
00:54:01,680 --> 00:54:04,600
Speaker 2: a coffee shop makeover. But it's clear there's lots of

1007
00:54:04,640 --> 00:54:09,200
Speaker 2: opportunities here to increase speed and efficiency, to achieve your objectives,

1008
00:54:09,360 --> 00:54:12,879
Speaker 2: and to dream beyond the current applications for this technology.

1009
00:54:13,840 --> 00:54:16,360
Speaker 2: At the end of the day, the scaling of AI

1010
00:54:16,480 --> 00:54:19,840
Speaker 2: will rely on the right infrastructure to support it. With

1011
00:54:19,920 --> 00:54:23,040
Speaker 2: the right tools, you can solve problems that are unique

1012
00:54:23,080 --> 00:54:37,920
Speaker 2: tier industry and improve the experience for your customers. Smart

1013
00:54:37,920 --> 00:54:41,200
Speaker 2: Talks with IBM is produced by Matt Romano, Amy Gains

1014
00:54:41,280 --> 00:54:45,640
Speaker 2: McQuaid and Jacob Goldstein. Were edited by Lydia gene Kott,

1015
00:54:46,000 --> 00:54:51,600
Speaker 2: mastering by Jake Koorsky. Theme song by Gramoscope. Special thanks

1016
00:54:51,600 --> 00:54:54,160
Speaker 2: to the eight Bar and IBM teams, as well as

1017
00:54:54,160 --> 00:54:58,080
Speaker 2: the Pushkin marketing team. Smart Talks with IBM is a

1018
00:54:58,080 --> 00:55:04,440
Speaker 2: production of Pushkin Industries and Ruby Studio at iHeartMedia. To

1019
00:55:04,480 --> 00:55:09,880
Speaker 2: find more Pushkin podcasts, listen on the iHeartRadio app, Apple Podcasts,

1020
00:55:10,000 --> 00:55:19,359
Speaker 2: or wherever you listen to podcasts. I'm Malcolm Gladwell. This

1021
00:55:19,520 --> 00:55:23,080
Speaker 2: is a paid advertisement from IBM. The conversations on this

1022
00:55:23,120 --> 00:55:40,560
Speaker 2: podcast don't necessarily represent IBM's positions, strategies or opinions.