1
00:00:03,120 --> 00:00:06,519
Speaker 1: Hello, Hello, Welcome to Smart Talks with IBM, a podcast

2
00:00:06,800 --> 00:00:12,440
Speaker 1: from Pushkin Industries, iHeartRadio and IBM. I'm Malcolm Godwell. This season,

3
00:00:12,480 --> 00:00:15,480
Speaker 1: we're diving back into the world of artificial intelligence, but

4
00:00:15,560 --> 00:00:21,160
Speaker 1: with a focus on the powerful concept of open its possibilities, implications,

5
00:00:21,440 --> 00:00:26,119
Speaker 1: and misconceptions. On today's episode, our season finale, I'm joined

6
00:00:26,120 --> 00:00:30,400
Speaker 1: by Rick Lewis, the senior vice president of Infrastructure at IBM.

7
00:00:31,000 --> 00:00:34,479
Speaker 1: Rick has had a remarkable career focused around product innovation.

8
00:00:35,200 --> 00:00:38,159
Speaker 1: He was actually a few years into retirement when IBM

9
00:00:38,240 --> 00:00:43,640
Speaker 1: came calling with an opportunity he just couldn't turn down. Thankfully,

10
00:00:43,840 --> 00:00:47,159
Speaker 1: Rick came out of retirement and today he oversees a

11
00:00:47,240 --> 00:00:52,720
Speaker 1: vast portfolio from storage and software to global customer support operations,

12
00:00:53,120 --> 00:00:55,440
Speaker 1: and he's engaged in one of the key problems facing

13
00:00:55,480 --> 00:01:00,400
Speaker 1: companies today, an explosion of data. In talking with Rick,

14
00:01:00,480 --> 00:01:02,800
Speaker 1: I can see that this problem of having so much

15
00:01:02,920 --> 00:01:07,000
Speaker 1: data is also an incredible opportunity because if you're able

16
00:01:07,040 --> 00:01:09,720
Speaker 1: to leverage that data to get the most value out

17
00:01:09,720 --> 00:01:12,119
Speaker 1: of it, then you can use it to help bring

18
00:01:12,160 --> 00:01:15,959
Speaker 1: your business into the future. We talked about the serious

19
00:01:15,959 --> 00:01:18,840
Speaker 1: computing power needed to scale AI, as well as the

20
00:01:18,840 --> 00:01:23,039
Speaker 1: ways that infrastructure storage solutions can be essential to enabling

21
00:01:23,120 --> 00:01:27,640
Speaker 1: this new world of possibilities. It's a really great conversation,

22
00:01:28,160 --> 00:01:34,720
Speaker 1: so let's get to it. I'm here with Rick Lewis. Rick,

23
00:01:34,959 --> 00:01:39,200
Speaker 1: Welcome than here. We are in the IBM's New York

24
00:01:39,240 --> 00:01:42,679
Speaker 1: City headquarters at one Madison Avenue. I'm going to start

25
00:01:42,680 --> 00:01:44,640
Speaker 1: with you're a hardware guy.

26
00:01:44,800 --> 00:01:48,400
Speaker 2: I'm a hardware guy. I grew up doing hardware chip engineering.

27
00:01:48,840 --> 00:01:51,320
Speaker 2: But like I tell a lot of people, a chip

28
00:01:51,360 --> 00:01:54,920
Speaker 2: engineering project is actually a giant software project with a

29
00:01:55,040 --> 00:01:57,480
Speaker 2: piece of hardware at the end of the project. Wise,

30
00:01:57,600 --> 00:02:00,440
Speaker 2: I think if you have that analytical brain like to

31
00:02:00,560 --> 00:02:04,200
Speaker 2: solve problems, you'd like to get things working, you can

32
00:02:04,240 --> 00:02:04,680
Speaker 2: do that in.

33
00:02:04,680 --> 00:02:08,800
Speaker 1: Soet But as being someone coming from a hardware background

34
00:02:08,840 --> 00:02:11,600
Speaker 1: mean that you think about problems in a different way.

35
00:02:12,240 --> 00:02:15,880
Speaker 2: I think one thing that you do from a hardware background,

36
00:02:15,880 --> 00:02:19,080
Speaker 2: and especially a chip background, is a chip spin and

37
00:02:19,120 --> 00:02:23,000
Speaker 2: costs millions of dollars, So you're a lot more likely

38
00:02:23,160 --> 00:02:26,200
Speaker 2: to make sure everything has a great chance of being

39
00:02:26,240 --> 00:02:28,720
Speaker 2: perfect from the get go. Or if you start kind

40
00:02:28,720 --> 00:02:31,720
Speaker 2: of from a software background, your general mindset is I

41
00:02:31,760 --> 00:02:33,400
Speaker 2: don't know, try this, see if it works. I don't know,

42
00:02:33,440 --> 00:02:35,840
Speaker 2: try that if it work, and you kind of iterated,

43
00:02:35,960 --> 00:02:39,440
Speaker 2: iterate chip people are a little more uptight about Okay,

44
00:02:39,480 --> 00:02:42,600
Speaker 2: if this first round of the chip breaks, costs us

45
00:02:42,639 --> 00:02:44,720
Speaker 2: for building another new round of the chip.

46
00:02:44,840 --> 00:02:47,919
Speaker 1: Yeah, so you're a little more You guys are spend

47
00:02:47,960 --> 00:02:53,840
Speaker 1: more time planning and planning verifying, tons of time verifying.

48
00:02:54,960 --> 00:02:58,960
Speaker 1: So you began your career as you look backward, yes, correct,

49
00:02:58,960 --> 00:03:00,000
Speaker 1: and you were there for how many years?

50
00:03:00,160 --> 00:03:01,639
Speaker 2: I was there for thirty two years?

51
00:03:02,160 --> 00:03:05,480
Speaker 1: Yes, And your last job there was I was leading.

52
00:03:05,160 --> 00:03:07,840
Speaker 2: The software defining cloud business. I had grown up a

53
00:03:07,880 --> 00:03:13,360
Speaker 2: hardware guy. I had done all kinds of hardware projects,

54
00:03:13,400 --> 00:03:17,480
Speaker 2: big complicated Unix servers and things like that, and then

55
00:03:18,280 --> 00:03:20,079
Speaker 2: you know, grew out of R and D and more

56
00:03:20,080 --> 00:03:24,840
Speaker 2: into the business realm, and then I'm much an innovator

57
00:03:24,880 --> 00:03:29,320
Speaker 2: at heart. I really like innovating new concepts things like that.

58
00:03:29,600 --> 00:03:32,800
Speaker 2: And what I learned is I enjoyed innovating business models

59
00:03:32,840 --> 00:03:35,960
Speaker 2: and software projects as much as I did hardware products

60
00:03:36,760 --> 00:03:40,480
Speaker 2: and projects, and so getting teams inspired toward doing that

61
00:03:40,880 --> 00:03:43,240
Speaker 2: was really a deep fascination for me. So I ended

62
00:03:43,320 --> 00:03:47,080
Speaker 2: up doing a fantastic variety of experiences and had a

63
00:03:47,080 --> 00:03:51,760
Speaker 2: successful run and honestly retired, intending to retire and do

64
00:03:51,840 --> 00:03:54,000
Speaker 2: some of my outside activities and things like that.

65
00:03:54,520 --> 00:03:56,920
Speaker 1: And then how long did you stay retired before IBM

66
00:03:56,960 --> 00:03:57,400
Speaker 1: can close?

67
00:03:57,680 --> 00:04:02,120
Speaker 2: Almost two years? And when when I first got at ALL,

68
00:04:02,360 --> 00:04:05,400
Speaker 2: I thought, no, I'm having too much fun. But I

69
00:04:05,400 --> 00:04:10,000
Speaker 2: would say three things really got me thinking hard about it.

70
00:04:10,080 --> 00:04:13,760
Speaker 2: One the industry that we're in, the IT industry. I

71
00:04:13,800 --> 00:04:15,800
Speaker 2: think it's the golden age. And what I mean by

72
00:04:15,880 --> 00:04:19,560
Speaker 2: that is for twenty years of that career, it is

73
00:04:19,640 --> 00:04:21,839
Speaker 2: kind of in the back office, say make sure that

74
00:04:21,880 --> 00:04:25,159
Speaker 2: stuff doesn't crash, and can you please reduce the cost

75
00:04:25,200 --> 00:04:27,240
Speaker 2: as much as possible, because it's not that important to

76
00:04:27,279 --> 00:04:30,719
Speaker 2: the main business. It's just a back office function. You

77
00:04:30,760 --> 00:04:33,080
Speaker 2: can see it right now. It is at the forefront

78
00:04:33,080 --> 00:04:37,359
Speaker 2: of all business revolution. It happened with the Internet. It

79
00:04:37,440 --> 00:04:41,120
Speaker 2: happened again with cloud and how that changed every ounce

80
00:04:41,160 --> 00:04:44,160
Speaker 2: of business, not just IT business, but all business. And

81
00:04:44,240 --> 00:04:47,359
Speaker 2: I think it's happening again with AI. So to be

82
00:04:47,480 --> 00:04:50,080
Speaker 2: in that career that long and to miss the kind

83
00:04:50,080 --> 00:04:52,560
Speaker 2: of this age where it's like this is front and center.

84
00:04:52,640 --> 00:04:56,320
Speaker 2: This changes everything about all businesses, not just technology businesses.

85
00:04:57,040 --> 00:05:00,720
Speaker 2: I was kind of feeling like gosh, you you trained

86
00:05:00,760 --> 00:05:05,400
Speaker 2: to be in these really awesome environments. Why wouldn't you

87
00:05:05,760 --> 00:05:08,080
Speaker 2: do that for a little while longer while you still

88
00:05:08,080 --> 00:05:11,880
Speaker 2: can do it. That combined with IBM and IBM seeing

89
00:05:11,920 --> 00:05:16,160
Speaker 2: the talent pool, the brilliant people at IBM, I worked

90
00:05:16,200 --> 00:05:18,640
Speaker 2: with a ton of brilliant people before I saw a

91
00:05:18,720 --> 00:05:21,640
Speaker 2: chance to work with even a larger staff of brilliant people.

92
00:05:21,760 --> 00:05:24,960
Speaker 2: And then the assets that IBM had, which is, you know,

93
00:05:25,000 --> 00:05:27,720
Speaker 2: they'd already been doing a lot of experimentation in AI,

94
00:05:27,880 --> 00:05:33,040
Speaker 2: they're working in quantum, the deep, rich heritage of successful projects.

95
00:05:33,040 --> 00:05:35,039
Speaker 2: I thought, who wouldn't want to kind of see if

96
00:05:35,080 --> 00:05:38,240
Speaker 2: they could be part of that next great wave of IBM.

97
00:05:38,320 --> 00:05:40,839
Speaker 2: And so I kind of decided, all right, I'm going

98
00:05:40,880 --> 00:05:42,920
Speaker 2: to put the outside interest on hold for a while

99
00:05:42,960 --> 00:05:43,960
Speaker 2: and get back in the game.

100
00:05:44,080 --> 00:05:47,400
Speaker 1: Along between the phone call, the first phone call and

101
00:05:47,480 --> 00:05:48,440
Speaker 1: you say, yes.

102
00:05:48,240 --> 00:05:52,480
Speaker 2: It was a while, It was probably six months. Arvind's

103
00:05:52,560 --> 00:05:55,160
Speaker 2: our CEO, teases me about that a lot. Yeah, he

104
00:05:55,279 --> 00:05:57,080
Speaker 2: was like, I don't think six months is that long?

105
00:05:57,160 --> 00:06:00,360
Speaker 2: It took a while if you were in retirement. I know. Yeah,

106
00:06:00,080 --> 00:06:03,240
Speaker 2: It's one thing to compare I'm working here and doing

107
00:06:03,240 --> 00:06:06,000
Speaker 2: this stuff versus working there, it's really hard to compare.

108
00:06:06,279 --> 00:06:08,800
Speaker 2: I'm doing exactly as I want to do every single

109
00:06:08,880 --> 00:06:10,720
Speaker 2: day when I wake up, and now I'm not going

110
00:06:10,760 --> 00:06:13,520
Speaker 2: to get to do that again. It took a while

111
00:06:13,520 --> 00:06:15,840
Speaker 2: for me to get over and I thought, I can't

112
00:06:15,880 --> 00:06:20,200
Speaker 2: miss this wave, and I'm really really happy that I did,

113
00:06:20,320 --> 00:06:23,960
Speaker 2: because we're doing some amazing, fun things and I'm getting

114
00:06:24,040 --> 00:06:26,520
Speaker 2: challenged in ways that I never did, so it's really fun.

115
00:06:26,800 --> 00:06:29,800
Speaker 1: Talk a little bit about your job here at IBM.

116
00:06:30,000 --> 00:06:34,200
Speaker 1: You oversee a kind of massive portfolio.

117
00:06:34,279 --> 00:06:37,200
Speaker 2: It's a big group, so I run the Infrastructure organization.

118
00:06:37,240 --> 00:06:40,640
Speaker 2: There's three main groups of products at IBM. There's the

119
00:06:40,640 --> 00:06:44,360
Speaker 2: Infrastructure group, which I run, the software group, and the

120
00:06:44,400 --> 00:06:49,600
Speaker 2: consulting group. And infrastructure is built up of mainframes, which

121
00:06:49,640 --> 00:06:53,800
Speaker 2: is called our Z portfolio, our servers which is our

122
00:06:53,839 --> 00:06:58,160
Speaker 2: power portfolio storage. By the way, those businesses include the

123
00:06:58,200 --> 00:07:00,520
Speaker 2: supply chain to build all of that stuff, so that's

124
00:07:00,520 --> 00:07:03,680
Speaker 2: in the group. Then I have the worldwide Customer Support

125
00:07:03,800 --> 00:07:08,080
Speaker 2: Organization it's called TLS Technology life Cycle Services, which is

126
00:07:08,120 --> 00:07:11,480
Speaker 2: a network of about thirteen thousand people around the globe

127
00:07:11,520 --> 00:07:13,600
Speaker 2: that make sure that everything runs and works when you

128
00:07:13,640 --> 00:07:17,320
Speaker 2: buy IBM products. And then also our IBM Cloud, which

129
00:07:17,360 --> 00:07:22,240
Speaker 2: is how we host applications and deliver as a service

130
00:07:22,280 --> 00:07:25,320
Speaker 2: products for our client based. So there's a lot. I

131
00:07:25,320 --> 00:07:27,280
Speaker 2: think it's about forty five thousand people total.

132
00:07:27,680 --> 00:07:32,240
Speaker 1: Do those components of the Infrastructure group are they aligned

133
00:07:32,240 --> 00:07:35,840
Speaker 1: in their trajectory or are they on different paths? And

134
00:07:36,120 --> 00:07:39,120
Speaker 1: I'm just curious what so navigattle of both. It's interesting

135
00:07:39,320 --> 00:07:42,720
Speaker 1: you would ask that because I think of all of

136
00:07:41,840 --> 00:07:45,840
Speaker 1: the challenges coming to the new company, there were things

137
00:07:45,880 --> 00:07:49,920
Speaker 1: I expected, things that they didn't expect. But getting that

138
00:07:50,080 --> 00:07:54,680
Speaker 1: culture right in that group has been a big challenge.

139
00:07:55,360 --> 00:08:01,720
Speaker 2: IBM has a great culture toward quality products, toward emphasizing

140
00:08:01,800 --> 00:08:04,360
Speaker 2: passion for the client and making sure that the client

141
00:08:04,480 --> 00:08:07,960
Speaker 2: is happy, and for delivering innovation on a scale that

142
00:08:08,760 --> 00:08:11,480
Speaker 2: you know, for more than one hundred years has been

143
00:08:12,280 --> 00:08:16,320
Speaker 2: extremely powerful. But with success comes some challenges. And with

144
00:08:16,400 --> 00:08:20,520
Speaker 2: that success you can tend to get a little bit insular,

145
00:08:20,720 --> 00:08:22,920
Speaker 2: like you don't keep an eye on the competition as well,

146
00:08:23,000 --> 00:08:25,520
Speaker 2: you can get more siloed, where you know, this is

147
00:08:25,600 --> 00:08:28,520
Speaker 2: my business unit, this is my business unit, I compete

148
00:08:28,600 --> 00:08:31,000
Speaker 2: with the other business unit. That's not a good thing

149
00:08:31,040 --> 00:08:35,480
Speaker 2: when you're a company, and you can get really risk averse,

150
00:08:35,880 --> 00:08:39,080
Speaker 2: meaning you feel like, hey, this is a successful business.

151
00:08:39,120 --> 00:08:40,839
Speaker 2: I don't want to do anything to mess it up,

152
00:08:41,160 --> 00:08:42,800
Speaker 2: so I don't need to try new things. Well, that's

153
00:08:42,840 --> 00:08:46,080
Speaker 2: exactly the recipe to kind of be shrinking, and infrastructure

154
00:08:46,120 --> 00:08:48,400
Speaker 2: had been shrinking for a little while, and so a

155
00:08:48,440 --> 00:08:52,040
Speaker 2: lot of what the challenge was for me was to

156
00:08:52,080 --> 00:08:56,280
Speaker 2: invigorate that risk taking and get to a growth mindset

157
00:08:56,280 --> 00:08:58,920
Speaker 2: where you're trying new things and seeing what works and

158
00:08:58,960 --> 00:09:02,040
Speaker 2: what doesn't work. Changing some of the models, like investing

159
00:09:02,080 --> 00:09:04,920
Speaker 2: a little bit less in hardware for some software differentiation

160
00:09:05,080 --> 00:09:08,640
Speaker 2: that goes into the hardware. So it's been very successful

161
00:09:08,679 --> 00:09:10,880
Speaker 2: so far, and it's been a good journey. It's almost

162
00:09:10,960 --> 00:09:11,720
Speaker 2: four years now.

163
00:09:12,120 --> 00:09:15,040
Speaker 1: Give me an example of what was a really hard

164
00:09:15,080 --> 00:09:16,800
Speaker 1: problem that you've dealt with in those sway years.

165
00:09:16,800 --> 00:09:19,720
Speaker 2: So, boy, a really hard problem?

166
00:09:19,800 --> 00:09:22,800
Speaker 1: An interesting and are you interesting is a better word

167
00:09:22,800 --> 00:09:23,200
Speaker 1: than ours.

168
00:09:23,400 --> 00:09:25,400
Speaker 2: One of the first things that I kind of chewed

169
00:09:25,440 --> 00:09:28,080
Speaker 2: on a little bit is I talked about how we

170
00:09:28,120 --> 00:09:31,240
Speaker 2: have Z power and storage. The Z and Power product

171
00:09:31,280 --> 00:09:34,040
Speaker 2: lines are well known in the industry. Is is really

172
00:09:34,080 --> 00:09:37,400
Speaker 2: fit for purpose computing that have strengths that you know

173
00:09:37,559 --> 00:09:41,439
Speaker 2: Z runs, you know most of the world's economic backbone.

174
00:09:41,480 --> 00:09:45,080
Speaker 2: And if you use a credit card, ninety percent of

175
00:09:45,120 --> 00:09:48,599
Speaker 2: credit card transactions for the globe go through these Z

176
00:09:49,120 --> 00:09:51,640
Speaker 2: mainframes there in every bank there. You know, it's a

177
00:09:51,640 --> 00:09:54,440
Speaker 2: big business. It's well known in the industry. Same with power,

178
00:09:54,880 --> 00:09:59,800
Speaker 2: very tuned and optimized for smaller operations than our giants

179
00:09:59,880 --> 00:10:05,880
Speaker 2: Z mainframes, but really mission critical workloads for retail, for insurance,

180
00:10:05,920 --> 00:10:09,080
Speaker 2: for banking, for all of that. Our storage business not

181
00:10:09,240 --> 00:10:12,000
Speaker 2: so well known. In fact, when I came I thought,

182
00:10:12,280 --> 00:10:14,680
Speaker 2: did they have storage? Well, I have storage when I

183
00:10:14,800 --> 00:10:17,240
Speaker 2: come in too. I so I got online and I thought,

184
00:10:17,320 --> 00:10:19,160
Speaker 2: it's still hard for me to tell did they have

185
00:10:19,240 --> 00:10:21,560
Speaker 2: storage or not? Now I own a storage business. So

186
00:10:21,640 --> 00:10:23,240
Speaker 2: one of the things was not just to get the

187
00:10:23,240 --> 00:10:26,839
Speaker 2: market perception up, but to invest in that business. Because

188
00:10:26,880 --> 00:10:29,559
Speaker 2: if you look at infrastructure overall around the globe, it's

189
00:10:29,600 --> 00:10:33,160
Speaker 2: growing at five percent a year. The infrastructure business had

190
00:10:33,200 --> 00:10:35,880
Speaker 2: been kind of flat to declining, and so a challenge

191
00:10:35,920 --> 00:10:37,840
Speaker 2: was how do we grab onto the growth. Well, one

192
00:10:37,880 --> 00:10:40,600
Speaker 2: of the biggest growth areas due to the explosion of

193
00:10:40,679 --> 00:10:43,280
Speaker 2: data in the world is storage, So what do you

194
00:10:43,320 --> 00:10:45,520
Speaker 2: do to kind of get on that growth rate. So

195
00:10:45,520 --> 00:10:49,160
Speaker 2: we did a lot of reinvigoration of the innovation in

196
00:10:49,200 --> 00:10:52,079
Speaker 2: that a lot of software value, add a lot of

197
00:10:52,400 --> 00:10:55,479
Speaker 2: doubling down on the things that are working. Portfolio rationalization

198
00:10:55,559 --> 00:10:57,760
Speaker 2: where you segment the market and you say, okay, we're

199
00:10:57,920 --> 00:10:59,680
Speaker 2: going to do less of this and really go big

200
00:10:59,720 --> 00:11:02,640
Speaker 2: in these areas, and that's been probably the most dramatic

201
00:11:02,640 --> 00:11:06,000
Speaker 2: turnaround inside the group. Is our storage thing. When you

202
00:11:06,040 --> 00:11:08,760
Speaker 2: say it's a hard problem, it's not just oh, you know,

203
00:11:08,800 --> 00:11:11,360
Speaker 2: how do we do the math? No, it's it's cultural.

204
00:11:11,520 --> 00:11:14,760
Speaker 2: It's strategy, and how do you get the strategy set.

205
00:11:14,800 --> 00:11:18,480
Speaker 2: It's segmentation, it's product strategy at a granular level across

206
00:11:18,480 --> 00:11:22,040
Speaker 2: a bunch of dimensions, and then putting the investment behind it.

207
00:11:22,040 --> 00:11:23,920
Speaker 2: It's a big challenge. It takes a long time, but

208
00:11:23,960 --> 00:11:25,280
Speaker 2: it's working, so we're happy with.

209
00:11:25,440 --> 00:11:27,760
Speaker 1: Yeah, tell me give me a little bit of perspective

210
00:11:27,960 --> 00:11:31,920
Speaker 1: on you've been there four years. Imagine we're having this

211
00:11:31,960 --> 00:11:36,080
Speaker 1: conversation four years ago. Yeah, what sorts of things have

212
00:11:36,120 --> 00:11:39,840
Speaker 1: happened over the last four years that have surprised you

213
00:11:39,920 --> 00:11:42,679
Speaker 1: that you didn't see come At least we had exactly

214
00:11:42,720 --> 00:11:44,160
Speaker 1: the same conversation four years ago.

215
00:11:45,559 --> 00:11:47,280
Speaker 2: No, because I didn't know what was in I'll tell

216
00:11:47,320 --> 00:11:51,120
Speaker 2: you some of the biggest surprises. I thought from the outside,

217
00:11:51,559 --> 00:11:54,720
Speaker 2: and you know, you hear from a lot of customers,

218
00:11:54,920 --> 00:11:58,040
Speaker 2: especially ten years ago, we're all going to cloud. We're

219
00:11:58,040 --> 00:11:59,640
Speaker 2: all doing it. So I thought, well, I wonder if

220
00:11:59,679 --> 00:12:02,560
Speaker 2: the main frame business is struggling. When I get inside

221
00:12:02,600 --> 00:12:04,920
Speaker 2: of there, I found the opposite to be true. The

222
00:12:04,960 --> 00:12:09,320
Speaker 2: mainframe business is actually flourishing because transaction demand across the

223
00:12:09,360 --> 00:12:12,679
Speaker 2: globe has done nothing but grow. And even more surprising

224
00:12:12,920 --> 00:12:15,360
Speaker 2: was the level of innovation that the team was already

225
00:12:15,440 --> 00:12:20,520
Speaker 2: doing in mainframes before I got here was astounding. For example,

226
00:12:20,720 --> 00:12:24,840
Speaker 2: we have AI. They were building AI technology into the

227
00:12:24,840 --> 00:12:29,680
Speaker 2: mainframe processors three years before chat GPT made everybody talk

228
00:12:29,760 --> 00:12:34,080
Speaker 2: about it in the industry, So that was really pleasantly surprising.

229
00:12:34,160 --> 00:12:38,960
Speaker 2: So that was wonderful. Other surprises. I knew about the

230
00:12:39,120 --> 00:12:42,079
Speaker 2: kind of the IP of IBM and the mystique in that,

231
00:12:42,440 --> 00:12:45,080
Speaker 2: and I used to joke with people, especially on the outside,

232
00:12:45,080 --> 00:12:46,520
Speaker 2: I said, I can't wait to get in there and

233
00:12:46,520 --> 00:12:49,360
Speaker 2: see what's in the big blue toolbox? Right, what are

234
00:12:49,400 --> 00:12:52,719
Speaker 2: all the things they have going on? I way underestimated

235
00:12:52,920 --> 00:12:55,280
Speaker 2: the size of the big blue toolbox and what was

236
00:12:55,320 --> 00:12:59,880
Speaker 2: in their meaning the amount of really hardcore research that

237
00:13:00,040 --> 00:13:03,480
Speaker 2: we're still doing into how to build chips and how

238
00:13:03,520 --> 00:13:06,600
Speaker 2: to get to things beyond two nanimeter and that kind

239
00:13:06,640 --> 00:13:12,600
Speaker 2: of capability packaging industry leading packaging technologies, and that's in

240
00:13:12,640 --> 00:13:15,439
Speaker 2: my hardware kind of patch quantum. The next thing that

241
00:13:15,480 --> 00:13:19,440
Speaker 2: will come after we're done talking about AI. You know,

242
00:13:20,080 --> 00:13:23,120
Speaker 2: all of those things were surprising, But it wasn't just that.

243
00:13:23,200 --> 00:13:25,760
Speaker 2: It was then the software innovations that are going on

244
00:13:25,920 --> 00:13:30,720
Speaker 2: heavy investment in AI technologies before it was really popular

245
00:13:30,760 --> 00:13:33,400
Speaker 2: to be talking about that. But as I saw that,

246
00:13:33,520 --> 00:13:36,120
Speaker 2: I thought this is going to be really fun because

247
00:13:36,160 --> 00:13:38,640
Speaker 2: I had a good feel for where the industry was going.

248
00:13:39,480 --> 00:13:42,120
Speaker 1: I just didn't and I knew, man, I know that.

249
00:13:42,120 --> 00:13:44,760
Speaker 2: Talent is really good and there's brilliant people there, but

250
00:13:44,800 --> 00:13:48,720
Speaker 2: I didn't know the level of IP frankly, that IBM

251
00:13:48,840 --> 00:13:51,760
Speaker 2: had at its disposal. And now you're seeing that in

252
00:13:51,800 --> 00:13:55,040
Speaker 2: things like Watson X and things like AI in mainframes,

253
00:13:55,040 --> 00:13:55,440
Speaker 2: et cetera.

254
00:13:55,840 --> 00:13:58,800
Speaker 1: Building on that, since you brought up AI, can you

255
00:13:58,840 --> 00:14:02,760
Speaker 1: walk me through what has to happen from your perspective,

256
00:14:02,760 --> 00:14:08,839
Speaker 1: from the infrastructure perspective to make the AI explosion work. Yeah,

257
00:14:08,880 --> 00:14:10,920
Speaker 1: so everyone wants to do more of this stuff. Yes,

258
00:14:11,080 --> 00:14:13,560
Speaker 1: clearly there has to be some underpinning of it.

259
00:14:14,000 --> 00:14:17,560
Speaker 2: Yeah, I would tell you, you know, I think that

260
00:14:17,720 --> 00:14:20,000
Speaker 2: people feel like where we're at right now in the

261
00:14:20,040 --> 00:14:23,120
Speaker 2: AI journey had to do with one specific piece of software.

262
00:14:23,480 --> 00:14:27,600
Speaker 2: I think the inflection point for that whole thing really

263
00:14:28,520 --> 00:14:32,240
Speaker 2: at its root was around hardware, meaning the algorithms needed

264
00:14:32,280 --> 00:14:34,880
Speaker 2: to do large language models. And all of that had

265
00:14:34,920 --> 00:14:37,560
Speaker 2: been around, they'd been talked about in the industry, but

266
00:14:37,600 --> 00:14:40,640
Speaker 2: at some point you hit a tipping point of hardware

267
00:14:40,720 --> 00:14:43,360
Speaker 2: capability where it's like, oh, now we can do this

268
00:14:43,440 --> 00:14:46,840
Speaker 2: in a broof force way, massive amounts of matrix math

269
00:14:46,920 --> 00:14:49,800
Speaker 2: to get weights correct so that you can do you know,

270
00:14:49,920 --> 00:14:53,240
Speaker 2: the right level of predictions that enable large language models.

271
00:14:53,600 --> 00:14:56,160
Speaker 2: And once we got to that horsepower, and that's why

272
00:14:56,200 --> 00:14:59,240
Speaker 2: you hear about giant GPUs that are driving this and

273
00:14:59,320 --> 00:15:01,600
Speaker 2: the sales of the et cetera. It's because we just

274
00:15:01,640 --> 00:15:03,560
Speaker 2: barely got over the hump where you can do these

275
00:15:03,600 --> 00:15:08,520
Speaker 2: big hard things in terms of hardware capability to do it.

276
00:15:09,000 --> 00:15:11,840
Speaker 1: Give me a layman, give me a sense of when

277
00:15:11,880 --> 00:15:14,040
Speaker 1: you say there was a kind of threshold where suddenly

278
00:15:14,040 --> 00:15:15,200
Speaker 1: these things became possible.

279
00:15:15,320 --> 00:15:18,120
Speaker 2: Yeah, I don't know if there's an exact number. But

280
00:15:19,280 --> 00:15:21,360
Speaker 2: and more basic question that I get from a lot

281
00:15:21,400 --> 00:15:24,080
Speaker 2: of people, you know, my friends and family outside, is

282
00:15:24,400 --> 00:15:29,120
Speaker 2: why GPUs. What does a GPU, it's graphics processor have

283
00:15:29,240 --> 00:15:33,640
Speaker 2: to do with AI. It's not, Well, graphics processors are

284
00:15:33,680 --> 00:15:37,440
Speaker 2: really good at this thing matrix math, because they're figuring

285
00:15:37,480 --> 00:15:40,720
Speaker 2: out how do I map a pixel? And and as

286
00:15:40,760 --> 00:15:44,840
Speaker 2: I move an object across the screen, it's essentially matrix

287
00:15:44,840 --> 00:15:47,000
Speaker 2: math to figure out, Okay, what does what does what

288
00:15:47,040 --> 00:15:49,640
Speaker 2: does this pixel on a screen look like? And what's

289
00:15:49,720 --> 00:15:52,600
Speaker 2: it's doing? And as you you know, we've gotten more

290
00:15:52,680 --> 00:15:56,400
Speaker 2: high resolution graphics, more high resolution monitors, et cetera. It's

291
00:15:56,440 --> 00:15:58,320
Speaker 2: a lot more pixels and a lot more math and

292
00:15:58,360 --> 00:16:00,760
Speaker 2: a lot more matrix math about how you compute that.

293
00:16:01,240 --> 00:16:03,560
Speaker 2: The first big thing that kind of started to look

294
00:16:03,680 --> 00:16:07,400
Speaker 2: like that, it turns out, was crypto and crypto mining,

295
00:16:07,440 --> 00:16:10,240
Speaker 2: and so you saw graphics companies starting to sell to crypto.

296
00:16:10,760 --> 00:16:12,840
Speaker 2: The technology got to a certain point and there were

297
00:16:12,920 --> 00:16:15,880
Speaker 2: use cases like bitcoin in that that kind of said, hey,

298
00:16:15,960 --> 00:16:17,840
Speaker 2: we need to do a lot of this matrix math

299
00:16:18,200 --> 00:16:20,960
Speaker 2: to be able to do that. So graphic chips were

300
00:16:20,960 --> 00:16:23,720
Speaker 2: a natural fit and that kind of sus same. But meanwhile,

301
00:16:23,760 --> 00:16:26,440
Speaker 2: behind the scenes, a lot of this ai AI is

302
00:16:26,480 --> 00:16:31,480
Speaker 2: about numeric calculations having to do with weights and matrices

303
00:16:31,480 --> 00:16:35,720
Speaker 2: that say, you know, giant consolidated things that predict what's

304
00:16:35,800 --> 00:16:38,200
Speaker 2: going to kind of happen based on what other things

305
00:16:38,200 --> 00:16:41,680
Speaker 2: have happened, just like predicting where pixel goes. But it's

306
00:16:41,760 --> 00:16:45,520
Speaker 2: really about being able to do enough data in jest

307
00:16:45,920 --> 00:16:48,320
Speaker 2: to be able to do and then the calculations to

308
00:16:48,360 --> 00:16:52,480
Speaker 2: be able to simplify things like entire sets of language

309
00:16:52,560 --> 00:16:55,920
Speaker 2: or giant chunks of the Internet, to get enough weightings

310
00:16:55,920 --> 00:16:58,120
Speaker 2: in there to be able to say, okay, we can

311
00:16:58,160 --> 00:17:01,760
Speaker 2: predict what you would say in this language based on

312
00:17:02,000 --> 00:17:04,560
Speaker 2: all of the volumes of stuff that we've seen that

313
00:17:04,600 --> 00:17:06,919
Speaker 2: when you start talking like this, the next word is

314
00:17:07,000 --> 00:17:08,600
Speaker 2: likely oh it's this yeah.

315
00:17:08,640 --> 00:17:10,919
Speaker 1: So, But my point is to get to that point.

316
00:17:11,000 --> 00:17:14,800
Speaker 1: That's threshold. We got there because was it because we

317
00:17:14,880 --> 00:17:17,719
Speaker 1: simply threw a lot more resources at the problem or

318
00:17:17,800 --> 00:17:22,040
Speaker 1: is it because the underlying technology got suddenly or gradually

319
00:17:22,200 --> 00:17:23,360
Speaker 1: so much more efficient.

320
00:17:23,480 --> 00:17:26,720
Speaker 2: It's always yes and yes. But you know, the industry

321
00:17:26,800 --> 00:17:29,040
Speaker 2: for a lot of years would talk about Moore's.

322
00:17:28,760 --> 00:17:33,080
Speaker 1: Law, Well, quick, will you define for us Moore's law

323
00:17:33,119 --> 00:17:34,800
Speaker 1: for those of those who's forgotten it.

324
00:17:35,040 --> 00:17:38,200
Speaker 2: Yeah, So Gordon Moore at Intel coined this thing. It

325
00:17:38,240 --> 00:17:42,960
Speaker 2: was basically that the horsepower I'm going to translate it

326
00:17:43,080 --> 00:17:48,680
Speaker 2: roughly of technology will double every couple of years. We're

327
00:17:48,720 --> 00:17:51,400
Speaker 2: still on Moore's law. More's law changed a little bit.

328
00:17:51,800 --> 00:17:54,640
Speaker 2: For a while. It was always about frequency. Things would

329
00:17:54,640 --> 00:17:57,800
Speaker 2: go faster, faster, faster. That kind of petered out. But

330
00:17:57,920 --> 00:18:01,120
Speaker 2: what happened is, rather than faster, faster, faster, we did

331
00:18:01,160 --> 00:18:04,399
Speaker 2: more and more and more. So rather than one operating

332
00:18:04,520 --> 00:18:08,360
Speaker 2: unit going a lot faster on its throughput, you put

333
00:18:08,400 --> 00:18:10,840
Speaker 2: ten operating units on a chip, Now you put one

334
00:18:10,920 --> 00:18:13,960
Speaker 2: hundred operating units on a chip, now a thousand. Some

335
00:18:14,040 --> 00:18:18,680
Speaker 2: of these problems, the matrix math problems scale parallel extremely well.

336
00:18:18,680 --> 00:18:21,000
Speaker 2: You don't have to do something really fast, you just

337
00:18:21,080 --> 00:18:23,080
Speaker 2: have to do a lot of the similar things in

338
00:18:23,119 --> 00:18:26,080
Speaker 2: parallel at the same time. So again that kind of

339
00:18:26,080 --> 00:18:28,840
Speaker 2: that extension of Moore's law, more and more hardware on

340
00:18:28,880 --> 00:18:30,280
Speaker 2: a chip to be able to do more and more

341
00:18:30,280 --> 00:18:33,520
Speaker 2: of those calculations in parallel and come up with it.

342
00:18:33,640 --> 00:18:36,760
Speaker 1: And we said, yeah, was that threshold predictable? In other words,

343
00:18:36,760 --> 00:18:39,639
Speaker 1: see people in the industry, Like you sit down X

344
00:18:39,680 --> 00:18:41,720
Speaker 1: number of years ago and say, when we get here,

345
00:18:42,720 --> 00:18:44,359
Speaker 1: AI is going to become much more.

346
00:18:44,200 --> 00:18:50,800
Speaker 2: Of a It's funny. The horsepower that very predictable, the

347
00:18:51,040 --> 00:18:54,520
Speaker 2: use cases not always so easy to kind of figure out.

348
00:18:54,600 --> 00:18:58,199
Speaker 2: That's where the human spirit kind of gets involved. I

349
00:18:58,240 --> 00:19:00,439
Speaker 2: think for some people that say, oh, I saw that coming,

350
00:19:00,720 --> 00:19:03,840
Speaker 2: But people have been predicting kind of the rise of

351
00:19:03,920 --> 00:19:06,840
Speaker 2: AI for twenty five years. Oh well, then when we

352
00:19:06,920 --> 00:19:08,800
Speaker 2: get to this next gener oh, when we get here,

353
00:19:08,960 --> 00:19:11,919
Speaker 2: it kind of hadn't happened. There's always a magic point

354
00:19:12,840 --> 00:19:15,159
Speaker 2: where you kind of get to where the technology and

355
00:19:15,240 --> 00:19:17,639
Speaker 2: the use case and somebody does something to kind of

356
00:19:17,680 --> 00:19:20,359
Speaker 2: make it catch on. And I think we're at one

357
00:19:20,400 --> 00:19:22,320
Speaker 2: of those moments in AI for sure right now. And

358
00:19:22,359 --> 00:19:24,560
Speaker 2: I don't think it's you know, people have said, oh,

359
00:19:24,560 --> 00:19:27,480
Speaker 2: this is just the latest wave of you know, I

360
00:19:27,840 --> 00:19:30,080
Speaker 2: hear I've heard this about a lot of technologies, but

361
00:19:30,400 --> 00:19:33,280
Speaker 2: AI is the technology the future, and it always will be.

362
00:19:33,560 --> 00:19:36,040
Speaker 2: I used to hear that you're not airing that now,

363
00:19:36,240 --> 00:19:39,840
Speaker 2: right It's like, no, it's primetime. It will change everything,

364
00:19:40,040 --> 00:19:42,320
Speaker 2: just like some of these other things changed everything.

365
00:19:42,560 --> 00:19:46,960
Speaker 1: I noticed it if personally, when I speak somewhere or

366
00:19:47,040 --> 00:19:50,640
Speaker 1: I'm listening to an audience somewhere. Over the last let's

367
00:19:50,680 --> 00:19:55,720
Speaker 1: say twelve months, there's always a whole bunch of AI questions.

368
00:19:55,960 --> 00:19:56,160
Speaker 2: Yes.

369
00:19:56,320 --> 00:19:58,520
Speaker 1: And if I go back two years ago, there were

370
00:19:58,520 --> 00:19:59,439
Speaker 1: no AI questions.

371
00:19:59,480 --> 00:19:59,680
Speaker 2: Yes.

372
00:20:00,280 --> 00:20:02,600
Speaker 1: Now my question is, so there's been this explosion on

373
00:20:02,640 --> 00:20:07,040
Speaker 1: the in popular fascination with what's going on AI. It

374
00:20:07,119 --> 00:20:08,280
Speaker 1: seems like the last year.

375
00:20:08,400 --> 00:20:10,760
Speaker 2: I agree with you.

376
00:20:09,960 --> 00:20:16,000
Speaker 1: In your world, when did the explosion of conversation around

377
00:20:16,000 --> 00:20:16,480
Speaker 1: this start.

378
00:20:17,240 --> 00:20:26,520
Speaker 2: It's I love this question because IBM had a fairly

379
00:20:26,680 --> 00:20:31,760
Speaker 2: big effort and business called Watson before Watson X. And

380
00:20:31,800 --> 00:20:34,720
Speaker 2: this is going back kind of ten years. I'll give

381
00:20:34,720 --> 00:20:37,560
Speaker 2: you another kind of example. I knew about a lot

382
00:20:37,600 --> 00:20:41,040
Speaker 2: of tablet technology before there was an iPad, a lot.

383
00:20:41,240 --> 00:20:43,600
Speaker 2: For ten years, there were a lot, but it kind

384
00:20:43,600 --> 00:20:47,280
Speaker 2: of takes a magic combination of the technology, the user experienced,

385
00:20:47,359 --> 00:20:50,000
Speaker 2: the software, and the need and the market ready for

386
00:20:50,040 --> 00:20:52,240
Speaker 2: it to kind of go. Now it's the thing. Now

387
00:20:52,280 --> 00:20:54,359
Speaker 2: we all have either an iPad or we have the

388
00:20:54,680 --> 00:20:57,600
Speaker 2: Google equivalent Tom and so I think this is a

389
00:20:57,680 --> 00:21:00,600
Speaker 2: little like that, meaning IBM was on the right track

390
00:21:00,680 --> 00:21:03,600
Speaker 2: with Watson. Some of the hardware wasn't there, the use

391
00:21:03,640 --> 00:21:06,280
Speaker 2: cases weren't exactly figured out. Some of the early use

392
00:21:06,320 --> 00:21:09,439
Speaker 2: cases didn't pan out perfectly. But the good news about

393
00:21:09,440 --> 00:21:12,919
Speaker 2: that is it's back to that culture of risk taking.

394
00:21:13,000 --> 00:21:15,480
Speaker 2: You don't look back on that and say, oh, we

395
00:21:15,480 --> 00:21:17,119
Speaker 2: shouldn't have done that, that was a bad idea. I know,

396
00:21:17,200 --> 00:21:19,040
Speaker 2: you look back on that and say, what did we learn?

397
00:21:19,320 --> 00:21:21,520
Speaker 2: How should we try something new? How would we pivot

398
00:21:21,560 --> 00:21:23,800
Speaker 2: this time. That's what we've done with Watson X, and

399
00:21:25,000 --> 00:21:27,760
Speaker 2: now that's a growing, healthy piece of our business and

400
00:21:27,920 --> 00:21:29,560
Speaker 2: very important our strategic picture.

401
00:21:29,600 --> 00:21:32,520
Speaker 1: So we're all in I just I've always investigated by

402
00:21:32,600 --> 00:21:38,560
Speaker 1: the gap between insider sense of what is happening an

403
00:21:38,560 --> 00:21:42,040
Speaker 1: outsider sense, like it absolutely is that in this case,

404
00:21:42,119 --> 00:21:45,520
Speaker 1: we've all been talking about and thinking about AI and

405
00:21:46,200 --> 00:21:48,520
Speaker 1: is it time for that and what does this mean,

406
00:21:48,560 --> 00:21:51,280
Speaker 1: et cetera. And yet none of us really predicted that

407
00:21:51,359 --> 00:21:54,760
Speaker 1: actual moment, which is kind of you know, early twenty

408
00:21:54,840 --> 00:21:58,199
Speaker 1: twenty two where it was like, oh, now you have

409
00:21:58,359 --> 00:22:02,480
Speaker 1: a simple, huge in the interface of software innovation combined

410
00:22:02,520 --> 00:22:08,760
Speaker 1: with large language models. There's a moment there where you're like, oh, Unlike,

411
00:22:09,040 --> 00:22:10,560
Speaker 1: you know, I think all of us are frustrated if

412
00:22:10,600 --> 00:22:12,879
Speaker 1: we ask our phone, hey tell me about this and

413
00:22:12,920 --> 00:22:15,359
Speaker 1: it says I found this on the web page.

414
00:22:15,400 --> 00:22:17,080
Speaker 2: That does you no good. But you know, all of

415
00:22:17,080 --> 00:22:21,200
Speaker 2: a sudden, with Chad GPT and some of these other things,

416
00:22:21,200 --> 00:22:22,879
Speaker 2: you could ask a question, it would give you a

417
00:22:22,880 --> 00:22:25,600
Speaker 2: clear answer. Sometimes is wrong, but at least it was

418
00:22:25,680 --> 00:22:27,880
Speaker 2: like I'm getting an answer rather than hey, I don't

419
00:22:27,920 --> 00:22:30,399
Speaker 2: know if there's some references. Good luck to you. And

420
00:22:30,440 --> 00:22:31,760
Speaker 2: that's really changing.

421
00:22:32,320 --> 00:22:36,840
Speaker 1: Talk about the kind of macro trends that are going

422
00:22:36,880 --> 00:22:39,919
Speaker 1: to shape your infrastructure battle.

423
00:22:40,040 --> 00:22:42,960
Speaker 2: Yeah, We've talked about if you already, but I'm actually

424
00:22:42,960 --> 00:22:46,879
Speaker 2: going to go a little different direction. So macro trans first.

425
00:22:47,359 --> 00:22:51,600
Speaker 2: And this one has been before even even this AI conversation,

426
00:22:51,680 --> 00:22:56,840
Speaker 2: that we've had explosion of data. As humans, we don't

427
00:22:57,000 --> 00:23:02,480
Speaker 2: think exponentially very well. Really struggle with exponential thinking. We

428
00:23:02,560 --> 00:23:05,000
Speaker 2: think linearly, Oh, there'll be more, there'll be more, they'll

429
00:23:05,040 --> 00:23:07,199
Speaker 2: be more, but we don't think well when it's like no,

430
00:23:07,320 --> 00:23:09,040
Speaker 2: they'll be more, and they'll be ten times more, and

431
00:23:09,080 --> 00:23:11,360
Speaker 2: then there'll be ten times that more. That's what's going

432
00:23:11,400 --> 00:23:14,359
Speaker 2: on with data right now in our industry. It's one

433
00:23:14,359 --> 00:23:16,760
Speaker 2: of the reasons that that storage business is doing so

434
00:23:16,840 --> 00:23:19,119
Speaker 2: well is there's just more and more and more data.

435
00:23:20,400 --> 00:23:22,240
Speaker 2: You know, you'd say, well, how can there be more data?

436
00:23:22,280 --> 00:23:24,679
Speaker 2: It's just life and that thing. The things that we

437
00:23:24,760 --> 00:23:28,320
Speaker 2: care about, video capture, video images, you know, the the

438
00:23:29,200 --> 00:23:31,879
Speaker 2: you I don't know from my parents, you needed a

439
00:23:32,000 --> 00:23:35,000
Speaker 2: drawer with all your family photos. Now we need gigabytes

440
00:23:35,000 --> 00:23:35,560
Speaker 2: and gigabytes.

441
00:23:35,640 --> 00:23:37,959
Speaker 1: You knew how many pictures my wife has taken off

442
00:23:37,960 --> 00:23:42,520
Speaker 1: our children, you would exactly, exactly, So that's your case. Now.

443
00:23:42,600 --> 00:23:45,240
Speaker 2: Think of companies who used to just think about their

444
00:23:45,280 --> 00:23:49,280
Speaker 2: transaction data. What's the ledger say that now have video

445
00:23:49,320 --> 00:23:52,440
Speaker 2: assets of all of their campaigns and their marketing. They're

446
00:23:52,480 --> 00:23:55,200
Speaker 2: trying to figure out, you know, what campaigns are working

447
00:23:55,200 --> 00:23:57,919
Speaker 2: the best. So it's just an explosion of data and

448
00:23:57,960 --> 00:24:01,760
Speaker 2: that's not going to stop with that, and more importantly,

449
00:24:01,840 --> 00:24:07,320
Speaker 2: getting value from that data is a massive trend in

450
00:24:07,359 --> 00:24:11,520
Speaker 2: the industry. Second trend AI, and this is the AI.

451
00:24:11,720 --> 00:24:13,760
Speaker 2: Not like we were just talking about about how it

452
00:24:13,840 --> 00:24:16,040
Speaker 2: changes how I search for things or how I learn

453
00:24:16,119 --> 00:24:20,480
Speaker 2: about things. But I would argue, dealing with that data,

454
00:24:20,560 --> 00:24:23,080
Speaker 2: how do I figure out what's in all those video streams?

455
00:24:23,119 --> 00:24:25,840
Speaker 2: How do I figure out Okay, I want all of

456
00:24:25,880 --> 00:24:28,880
Speaker 2: the chunks of my corporate video that have to do

457
00:24:28,960 --> 00:24:33,000
Speaker 2: with client buying some specific product or something. That's a

458
00:24:33,640 --> 00:24:35,959
Speaker 2: different problem. It's not just okay, we'll look it up

459
00:24:35,960 --> 00:24:39,560
Speaker 2: in a spreadsheet and here's the math associated with that.

460
00:24:39,560 --> 00:24:41,920
Speaker 2: That is a huge trend in the industry. You're seeing

461
00:24:41,920 --> 00:24:44,159
Speaker 2: it play out in this regard. It's a little different

462
00:24:44,240 --> 00:24:48,240
Speaker 2: bent on AI. Fraud detection is the one that we

463
00:24:48,560 --> 00:24:51,439
Speaker 2: cite in our mainframes. It's a similar problem where it

464
00:24:51,600 --> 00:24:54,359
Speaker 2: was kind of a traditional AI problem. Look up a rule.

465
00:24:54,920 --> 00:24:58,399
Speaker 2: You know, if somebody does two small transactions, then a

466
00:24:58,440 --> 00:25:00,600
Speaker 2: massive one it might be fraud, right because they were

467
00:25:00,640 --> 00:25:04,280
Speaker 2: seeing whether it were now to detect fraud, you might

468
00:25:04,320 --> 00:25:08,680
Speaker 2: be saying, okay to transactions then a huge one. Plus

469
00:25:08,760 --> 00:25:12,159
Speaker 2: does this entity have a real address? Second, is there

470
00:25:12,200 --> 00:25:15,640
Speaker 2: any web traffic on you know, better business bureau kind

471
00:25:15,680 --> 00:25:17,399
Speaker 2: of things that says this is a bad business that

472
00:25:17,760 --> 00:25:19,720
Speaker 2: can help you with fraud. So it's a lot more

473
00:25:19,760 --> 00:25:22,840
Speaker 2: of a it's an expotent still problem. It's a holistic

474
00:25:22,920 --> 00:25:25,639
Speaker 2: problem that it takes a lot more than just you know,

475
00:25:25,880 --> 00:25:28,720
Speaker 2: little chunks of rules, et cetera. And then the third

476
00:25:28,720 --> 00:25:32,600
Speaker 2: one you know after AI is the nature of hybrid

477
00:25:32,600 --> 00:25:36,240
Speaker 2: it or hybrid computing. For a while ten years ago

478
00:25:36,320 --> 00:25:39,520
Speaker 2: when cloud was on the rise, I think the notion

479
00:25:39,600 --> 00:25:42,800
Speaker 2: of hybrid computing basically having to do with things in

480
00:25:42,840 --> 00:25:46,840
Speaker 2: the cloud versus things that people still have on the

481
00:25:46,880 --> 00:25:50,680
Speaker 2: premises inside of business, it was almost a religious argument.

482
00:25:51,240 --> 00:25:54,199
Speaker 2: Now it's no, it's the reality. And the reason is

483
00:25:54,240 --> 00:25:57,919
Speaker 2: because that data that I talked about is the lifeblood

484
00:25:57,960 --> 00:26:01,840
Speaker 2: of these companies, particularly IBM's come. Companies are clients that

485
00:26:02,040 --> 00:26:05,240
Speaker 2: usually that data has to be secure, they have to

486
00:26:05,240 --> 00:26:07,879
Speaker 2: be able to get value from it. It is the

487
00:26:07,920 --> 00:26:09,960
Speaker 2: lifeblood of the company. If you go to an ATM

488
00:26:10,000 --> 00:26:13,720
Speaker 2: and you can't get your money out to our financial transactions,

489
00:26:14,480 --> 00:26:16,800
Speaker 2: if that lasts a day, you're probably going to change

490
00:26:16,800 --> 00:26:19,840
Speaker 2: banks immediately. So it's like life or death for these companies.

491
00:26:21,720 --> 00:26:26,160
Speaker 2: So having that hybrid infrastructure so that they can still

492
00:26:26,200 --> 00:26:29,640
Speaker 2: hold their data, you still interact with clouds and still

493
00:26:29,640 --> 00:26:32,600
Speaker 2: get value from it from AI, that's kind of the

494
00:26:32,720 --> 00:26:37,320
Speaker 2: magic where we play, and it's a huge business opportunity.

495
00:26:37,600 --> 00:26:41,239
Speaker 2: It is a true inflection point for the industry. I'm

496
00:26:41,280 --> 00:26:42,280
Speaker 2: going to go back.

497
00:26:43,359 --> 00:26:45,200
Speaker 1: I interrupted you when you were in the middle of

498
00:26:45,240 --> 00:26:48,119
Speaker 1: a rellion. We were talking about what has to happen

499
00:26:48,240 --> 00:26:53,640
Speaker 1: for AI to scale from the infrastructure standpoint. You gave

500
00:26:53,680 --> 00:26:55,959
Speaker 1: one example that I got you off on a tangent.

501
00:26:56,320 --> 00:26:59,840
Speaker 1: Can you go back and talk very so practically. So,

502
00:27:00,119 --> 00:27:03,000
Speaker 1: I'm I'm a big company. I have all these dreams

503
00:27:03,080 --> 00:27:06,280
Speaker 1: of AI, of how I'm going to use this dratically.

504
00:27:06,560 --> 00:27:09,720
Speaker 1: So give me a very granular sense of the works

505
00:27:09,880 --> 00:27:12,800
Speaker 1: you have to do, yeah to make that dream possible.

506
00:27:13,240 --> 00:27:16,479
Speaker 2: So let me first say what the company has to do,

507
00:27:16,520 --> 00:27:18,480
Speaker 2: and then maybe I'll say, then how do I help them?

508
00:27:18,560 --> 00:27:20,720
Speaker 2: If that makes sense? So if I'm a company and

509
00:27:20,760 --> 00:27:22,719
Speaker 2: I want to do that, So it turns out I

510
00:27:22,760 --> 00:27:26,560
Speaker 2: am a company meaning I want to use AI in

511
00:27:26,600 --> 00:27:30,439
Speaker 2: my processes. I mentioned that I have a global network

512
00:27:30,480 --> 00:27:34,399
Speaker 2: of thirteen thousand employees that support our infrastructure around the world.

513
00:27:35,040 --> 00:27:40,440
Speaker 2: That challenge is a great challenge for AI. That means

514
00:27:40,480 --> 00:27:45,040
Speaker 2: I have data for every customer situation for thirteen thousand

515
00:27:45,080 --> 00:27:48,520
Speaker 2: employees globally around the world on what was their problem,

516
00:27:48,600 --> 00:27:52,040
Speaker 2: how did we fix it, what next steps did they

517
00:27:52,040 --> 00:27:54,480
Speaker 2: have to do, how did they remediate that? That data

518
00:27:54,560 --> 00:27:57,040
Speaker 2: is extremely valuable to me because if I can get

519
00:27:57,080 --> 00:27:59,960
Speaker 2: better at doing that than anybody else in the world,

520
00:28:00,080 --> 00:28:02,320
Speaker 2: that brings my cost down. I sell more products, I

521
00:28:02,359 --> 00:28:05,239
Speaker 2: sell more service, I sell more anything. So what I

522
00:28:05,359 --> 00:28:07,280
Speaker 2: have to do to get there is I have to

523
00:28:07,280 --> 00:28:10,760
Speaker 2: figure out, Okay, what's my objective. I have a couple objectives. One,

524
00:28:10,840 --> 00:28:13,359
Speaker 2: I want customers to be able to support themselves without

525
00:28:13,359 --> 00:28:17,239
Speaker 2: even calling me, first off, and I don't want when

526
00:28:17,280 --> 00:28:19,719
Speaker 2: they call for the first answer to come back to

527
00:28:19,720 --> 00:28:23,080
Speaker 2: be did you try rebooting? Because I think that irritates

528
00:28:23,119 --> 00:28:25,520
Speaker 2: every single one of them. Did you try? Of course

529
00:28:25,520 --> 00:28:28,280
Speaker 2: I tried rebooting. I've had a laptop my entire of course.

530
00:28:28,320 --> 00:28:32,480
Speaker 2: I well, okay, well then tell me, okay, what firmware version,

531
00:28:32,520 --> 00:28:34,600
Speaker 2: all that other stuff. Okay, we know this interaction. So

532
00:28:35,880 --> 00:28:37,800
Speaker 2: that's kind of the problem set. Do I want that

533
00:28:37,960 --> 00:28:41,280
Speaker 2: to be customers solving their own problems? Well, even for

534
00:28:41,400 --> 00:28:43,840
Speaker 2: my support agents, I want something in their pocket on

535
00:28:43,880 --> 00:28:46,800
Speaker 2: their phone where they say I'm seeing these symptoms. It says, oh,

536
00:28:47,040 --> 00:28:49,760
Speaker 2: this happening around the globe. Here's kind of specific me.

537
00:28:49,880 --> 00:28:53,480
Speaker 2: So there's my problems. What does it mean for infrastructure

538
00:28:53,520 --> 00:28:57,360
Speaker 2: on the back end? So first I got to get

539
00:28:57,360 --> 00:29:00,280
Speaker 2: all that data together, right, all of those customers, all

540
00:29:00,320 --> 00:29:03,600
Speaker 2: that customer support around the globe, et cetera. That needs

541
00:29:03,600 --> 00:29:06,160
Speaker 2: to be stored. That's a big set of data and

542
00:29:06,200 --> 00:29:09,520
Speaker 2: some of it's not just fix and that kind of thing.

543
00:29:09,600 --> 00:29:12,040
Speaker 2: Some of it is, Okay, you know what was the

544
00:29:12,080 --> 00:29:14,680
Speaker 2: firmware version, who was the tech because it can matter.

545
00:29:15,200 --> 00:29:17,280
Speaker 2: Is this their first time fixing this problem? Is it

546
00:29:17,320 --> 00:29:19,600
Speaker 2: there one hundred and fiftieth time? What's their level? It's

547
00:29:19,640 --> 00:29:25,080
Speaker 2: a very complicated problem. Ingesting all that data takes an architecture.

548
00:29:25,120 --> 00:29:28,000
Speaker 2: We have a product called Scale, which is one of

549
00:29:28,000 --> 00:29:31,480
Speaker 2: our storage projects that actually makes it easy to ingest

550
00:29:31,480 --> 00:29:34,840
Speaker 2: all that data, get it organized, et cetera, and then

551
00:29:36,280 --> 00:29:38,880
Speaker 2: have a model. It's a whole different process to kind

552
00:29:38,880 --> 00:29:40,680
Speaker 2: of say did we train our model? We can train

553
00:29:40,720 --> 00:29:43,000
Speaker 2: our own models. Inside of IBM, we have a granite

554
00:29:43,040 --> 00:29:46,360
Speaker 2: set of models. Those models we fine tune and then

555
00:29:46,400 --> 00:29:48,800
Speaker 2: we inference based on those models. So we can do

556
00:29:48,840 --> 00:29:51,680
Speaker 2: that inferencing in our cloud. I have a cloud set

557
00:29:51,680 --> 00:29:54,200
Speaker 2: of infrastructure, or in my power servers. We can do

558
00:29:54,240 --> 00:29:59,080
Speaker 2: inferencing with our capabilities and say, okay, based on what

559
00:29:59,200 --> 00:30:01,920
Speaker 2: I'm saying, there's what the remediation that you should do

560
00:30:02,000 --> 00:30:05,280
Speaker 2: for that customer. We already are doing that today. We've

561
00:30:05,360 --> 00:30:10,920
Speaker 2: seen over a third of our support calls have had

562
00:30:11,080 --> 00:30:14,200
Speaker 2: significant reduction in the amount of time that it takes

563
00:30:14,200 --> 00:30:17,560
Speaker 2: to resolve that support call. Just by what I said

564
00:30:17,680 --> 00:30:18,360
Speaker 2: right there.

565
00:30:18,480 --> 00:30:22,040
Speaker 1: That I've really been curious about this if I had

566
00:30:22,080 --> 00:30:26,280
Speaker 1: reduced something like AI into that equation as you just did. Yeah,

567
00:30:26,320 --> 00:30:29,320
Speaker 1: and you said we've already seen a thirty percent Say

568
00:30:29,360 --> 00:30:30,719
Speaker 1: did you say thirty percent reduction?

569
00:30:31,080 --> 00:30:35,600
Speaker 2: Thirty percent of our interactions have seen significant reduction in

570
00:30:36,000 --> 00:30:36,600
Speaker 2: those time?

571
00:30:36,720 --> 00:30:39,720
Speaker 1: Was that your primary goal to reduce the time of

572
00:30:39,760 --> 00:30:42,520
Speaker 1: the interaction? And it was if you if everything else

573
00:30:42,600 --> 00:30:44,680
Speaker 1: was the same, but all but what you were doing

574
00:30:44,840 --> 00:30:45,960
Speaker 1: was shrinking the amount of time.

575
00:30:46,000 --> 00:30:48,760
Speaker 2: That would you want one of the primary goals, so

576
00:30:49,880 --> 00:30:53,560
Speaker 2: to us, in that business net promoter score kind of

577
00:30:53,560 --> 00:30:56,760
Speaker 2: the satisfaction of a client is the supreme goal. What

578
00:30:56,920 --> 00:31:01,000
Speaker 2: makes them satisfied doesn't cost me a fortune, happens really quickly,

579
00:31:01,200 --> 00:31:03,120
Speaker 2: and if I can do it myself, I'd be thrilled.

580
00:31:03,840 --> 00:31:06,560
Speaker 2: It affects all of those right. It kind of says

581
00:31:06,680 --> 00:31:09,040
Speaker 2: it got resolved faster, it didn't cost me an arm

582
00:31:09,040 --> 00:31:11,120
Speaker 2: and the leg because the deck was barely here, because

583
00:31:11,120 --> 00:31:14,160
Speaker 2: it's a common problem, or I solved it myself without

584
00:31:14,160 --> 00:31:17,920
Speaker 2: even calling, So all of those objectives would kind of

585
00:31:18,000 --> 00:31:19,960
Speaker 2: hit across all so that now you see it so

586
00:31:20,000 --> 00:31:22,440
Speaker 2: that's a little microcosm. That's just me and my customer

587
00:31:22,480 --> 00:31:25,920
Speaker 2: support business. Now, think of how many problems for businesses

588
00:31:25,960 --> 00:31:29,120
Speaker 2: around the world there are like that. It's not a

589
00:31:29,200 --> 00:31:32,160
Speaker 2: it's not like a new AI application that changes the

590
00:31:32,320 --> 00:31:37,160
Speaker 2: entire user experience. That's those will come, But right now

591
00:31:37,400 --> 00:31:39,760
Speaker 2: it's kind of practical, which is, I just want to

592
00:31:39,800 --> 00:31:42,720
Speaker 2: do what I'm doing better and faster, and I can

593
00:31:42,760 --> 00:31:45,240
Speaker 2: get immediate economic return from those things.

594
00:31:45,240 --> 00:31:48,320
Speaker 1: How long How long did it take you to just

595
00:31:48,400 --> 00:31:51,760
Speaker 1: stick with that example of the customer reduction reducing thirty

596
00:31:51,760 --> 00:31:54,440
Speaker 1: percent of the time? How long from the very beginning

597
00:31:54,480 --> 00:31:57,800
Speaker 1: of that project? Yeah to that thirty percent reduction was?

598
00:31:57,800 --> 00:31:58,200
Speaker 1: How long?

599
00:31:58,840 --> 00:32:04,200
Speaker 2: Less than a year? And yeah, So one of the challenges,

600
00:32:04,520 --> 00:32:08,160
Speaker 2: and this is interesting with a very large organization, as

601
00:32:08,200 --> 00:32:10,640
Speaker 2: you can imagine, just like you're seeing in the industry,

602
00:32:11,520 --> 00:32:15,120
Speaker 2: we don't have a problem of generating ideas for how

603
00:32:15,160 --> 00:32:18,800
Speaker 2: AI could help us. We actually have a problem filtering

604
00:32:19,200 --> 00:32:23,040
Speaker 2: the thousands of ideas from our employees and from everywhere.

605
00:32:23,080 --> 00:32:25,400
Speaker 2: It's like, hey, we could use AI to and filtering

606
00:32:25,440 --> 00:32:27,600
Speaker 2: down and saying, okay, which of these will have a

607
00:32:27,680 --> 00:32:31,440
Speaker 2: return on investment quickly and at a level that sustains

608
00:32:31,480 --> 00:32:34,200
Speaker 2: that's worth kind of going and investing in the infrastructure

609
00:32:34,200 --> 00:32:37,720
Speaker 2: and the software and kind of making that happen.

610
00:32:37,800 --> 00:32:41,120
Speaker 1: Is that unusual If I talked to you twenty five

611
00:32:41,200 --> 00:32:43,480
Speaker 1: years ago and said, do you have a problem with

612
00:32:43,560 --> 00:32:44,960
Speaker 1: too many good ideas or too few?

613
00:32:44,960 --> 00:32:51,120
Speaker 2: What was you said in this specific area? Probably too few,

614
00:32:51,280 --> 00:32:54,680
Speaker 2: because at some point you reach diminishing returns. So, for example,

615
00:32:54,760 --> 00:32:59,000
Speaker 2: let's use this same example. Can those thirteen thousand technicians

616
00:32:59,080 --> 00:33:03,440
Speaker 2: go faster? Can they spend less time driving to the side.

617
00:33:03,440 --> 00:33:05,160
Speaker 2: I mean, there's only so much you can kind of

618
00:33:05,200 --> 00:33:07,560
Speaker 2: do on those things. But if you can get them

619
00:33:07,560 --> 00:33:09,840
Speaker 2: an answer to the problem and maybe even avoid them

620
00:33:09,880 --> 00:33:12,440
Speaker 2: having to visit at all because the client helped themselves,

621
00:33:12,840 --> 00:33:15,840
Speaker 2: that's a step function. So that's why people are kind

622
00:33:15,840 --> 00:33:20,240
Speaker 2: of talking about there's a business revolution coming with AI

623
00:33:20,360 --> 00:33:23,360
Speaker 2: where there are some step function changes that can be there.

624
00:33:23,400 --> 00:33:26,760
Speaker 2: And notice I didn't say I'm going to have less

625
00:33:26,760 --> 00:33:30,320
Speaker 2: of those agents. That's not my objective. My objective and

626
00:33:30,400 --> 00:33:32,760
Speaker 2: I think that's the fear in the industry about AI

627
00:33:32,840 --> 00:33:35,200
Speaker 2: is going to eliminate all the jobs. No, I just

628
00:33:35,280 --> 00:33:39,240
Speaker 2: created thirteen thousand superpowered agents that can do more right

629
00:33:39,320 --> 00:33:41,520
Speaker 2: and so I'm not just going to support IBM products.

630
00:33:41,760 --> 00:33:43,920
Speaker 2: I'm going to go out and support other people's products

631
00:33:43,960 --> 00:33:45,600
Speaker 2: because I know how to do that really well. And

632
00:33:45,640 --> 00:33:48,400
Speaker 2: once I have the data on how to fix their problems,

633
00:33:48,840 --> 00:33:52,480
Speaker 2: I may just have a customer support business that's independent

634
00:33:52,520 --> 00:33:55,240
Speaker 2: of my boxes. So you know, I think that's where

635
00:33:55,280 --> 00:33:57,840
Speaker 2: people sometimes get it wrong. And the AI thing is

636
00:33:58,400 --> 00:34:04,400
Speaker 2: it's like, you know, word processing eliminate the need for writers. No,

637
00:34:04,560 --> 00:34:09,120
Speaker 2: it enabled writing instead of mucking around with mimeographic machines

638
00:34:09,120 --> 00:34:10,560
Speaker 2: and click and click typewriters.

639
00:34:10,560 --> 00:34:12,959
Speaker 1: It may have enabled too much writing? Yeah, maybe maybe

640
00:34:13,680 --> 00:34:16,839
Speaker 1: can I give you a hypothetical? Uh? And I asked

641
00:34:16,880 --> 00:34:18,920
Speaker 1: this because I read I was at some convers and

642
00:34:18,920 --> 00:34:21,040
Speaker 1: I ran to some guy from the I R S

643
00:34:22,160 --> 00:34:25,640
Speaker 1: who was really, really, really really excited about AI. So

644
00:34:25,719 --> 00:34:29,919
Speaker 1: let's suppose they call you up and they say, you're

645
00:34:29,920 --> 00:34:33,560
Speaker 1: going to talk to ask the I the IRKY. I

646
00:34:33,640 --> 00:34:38,560
Speaker 1: call you up and I say, Rick, Uh, clearly there's

647
00:34:38,560 --> 00:34:42,120
Speaker 1: something that we could do for the I R S

648
00:34:42,160 --> 00:34:44,280
Speaker 1: if we work together. Yeah, what would your answering?

649
00:34:45,560 --> 00:34:49,399
Speaker 2: Of course? No, I think we sell to a lot

650
00:34:49,440 --> 00:34:52,920
Speaker 2: of government agencies. I can imagine in the business that

651
00:34:52,960 --> 00:34:57,640
Speaker 2: we're in, we enable a lot of social security transactions

652
00:34:57,640 --> 00:35:01,600
Speaker 2: and things like that through our mainframes. And I think,

653
00:35:01,960 --> 00:35:05,080
Speaker 2: you know, we're in the business of helping whatever client

654
00:35:05,200 --> 00:35:07,120
Speaker 2: get the most out of their data and be able

655
00:35:07,160 --> 00:35:10,239
Speaker 2: to secure it and be able to do analytics with

656
00:35:10,600 --> 00:35:13,040
Speaker 2: and IRS has a heck of a lot of data,

657
00:35:13,160 --> 00:35:15,520
Speaker 2: So yes, we would help them. Do you know how

658
00:35:15,520 --> 00:35:17,520
Speaker 2: the amount of data they have compares to some of

659
00:35:17,560 --> 00:35:20,440
Speaker 2: the corporate clients you I don't know specifically for the

660
00:35:20,520 --> 00:35:22,680
Speaker 2: IRS how much data they have, but I would assume

661
00:35:22,719 --> 00:35:26,480
Speaker 2: it's a whole lot. It's mountains. But but that's our business.

662
00:35:26,600 --> 00:35:29,759
Speaker 2: I mean, it's interesting sometimes people of that what's the

663
00:35:29,880 --> 00:35:34,360
Speaker 2: most you know, what is it that that IBM has

664
00:35:34,480 --> 00:35:37,640
Speaker 2: that's of great value? Is it a server? Is it

665
00:35:38,000 --> 00:35:41,000
Speaker 2: a storage array? Is it you know, software and all that.

666
00:35:41,320 --> 00:35:45,520
Speaker 2: What we have is the most important entities in the

667
00:35:45,560 --> 00:35:49,120
Speaker 2: world have their data on our stuff. The most important

668
00:35:49,200 --> 00:35:52,680
Speaker 2: data in the world. It's not you know, pictures of

669
00:35:52,719 --> 00:35:55,399
Speaker 2: your grandkids and things like that. Generally for us, it's

670
00:35:55,440 --> 00:35:58,799
Speaker 2: all of the financial transactions that happen globally, right, It's

671
00:35:58,840 --> 00:36:01,520
Speaker 2: all of the it's the world's economy is kind of

672
00:36:01,560 --> 00:36:05,600
Speaker 2: running through our systems, and so we take that really seriously.

673
00:36:05,760 --> 00:36:08,600
Speaker 2: You know, you would be distraught if you lost one

674
00:36:08,600 --> 00:36:11,520
Speaker 2: photo on your laptop or whatever. But you know, if

675
00:36:11,520 --> 00:36:14,600
Speaker 2: we lose a transaction, like somebody moves a big amount

676
00:36:14,640 --> 00:36:17,440
Speaker 2: of money and it's like, well, don't know what happened there,

677
00:36:17,680 --> 00:36:21,160
Speaker 2: it is a massive deal, right, so that doesn't happen.

678
00:36:21,280 --> 00:36:22,840
Speaker 1: But I want to go back to my irs example

679
00:36:22,840 --> 00:36:26,720
Speaker 1: for us, Yes, so one, is it reasonable to assume

680
00:36:27,000 --> 00:36:32,040
Speaker 1: that you could that somebody IBM or somebody else could

681
00:36:32,080 --> 00:36:34,560
Speaker 1: in a short period of time put together not just

682
00:36:34,600 --> 00:36:40,160
Speaker 1: the AI capability to audit returns, but also this the

683
00:36:40,200 --> 00:36:43,439
Speaker 1: infrastructure support for that in a reasonable amount of time

684
00:36:43,440 --> 00:36:45,879
Speaker 1: for a reasonable amount of cost. Or is it over?

685
00:36:46,280 --> 00:36:49,120
Speaker 1: Is it going to the moon? Or is it it?

686
00:36:49,360 --> 00:36:53,200
Speaker 2: Definitely? I mean, so we're already doing that kind of

687
00:36:53,239 --> 00:36:58,520
Speaker 2: thing right across a network of banks and others, essentially

688
00:36:58,640 --> 00:37:02,560
Speaker 2: all credit card transactions for all of the world to

689
00:37:02,600 --> 00:37:05,839
Speaker 2: go through our systems, so that in some ways is

690
00:37:05,880 --> 00:37:09,400
Speaker 2: more volume than the datch returns of the US people.

691
00:37:09,480 --> 00:37:12,680
Speaker 2: And they're W two's and all that stuff, and we

692
00:37:12,760 --> 00:37:15,560
Speaker 2: do that stuff too. I try not to describe it

693
00:37:15,600 --> 00:37:17,799
Speaker 2: too much in detail, but we definitely do a lot

694
00:37:17,840 --> 00:37:22,799
Speaker 2: of that. In fact, I think most of what If

695
00:37:22,800 --> 00:37:26,200
Speaker 2: you think, okay, what is super critical data? Who would

696
00:37:26,200 --> 00:37:29,680
Speaker 2: be doing the business transaction processing? It is most likely

697
00:37:29,840 --> 00:37:33,440
Speaker 2: us in almost all cases, whether it's government things or

698
00:37:34,160 --> 00:37:37,239
Speaker 2: private or banks or that kind of thing. That's what

699
00:37:37,280 --> 00:37:37,600
Speaker 2: we do.

700
00:37:37,960 --> 00:37:39,920
Speaker 1: Rick We're going to end with the where we always

701
00:37:40,040 --> 00:37:41,880
Speaker 1: end with a couple of quick fire questions.

702
00:37:41,960 --> 00:37:42,960
Speaker 2: Okay, here we go.

703
00:37:43,840 --> 00:37:46,920
Speaker 1: What single piece of advice would you give to businesses

704
00:37:46,960 --> 00:37:50,280
Speaker 1: trying to use AI in an effective way? The simple

705
00:37:50,400 --> 00:37:54,200
Speaker 1: version is get started. By get started, I mean, think

706
00:37:54,239 --> 00:37:57,600
Speaker 1: of what is something that I want to improve. The

707
00:37:57,640 --> 00:37:59,759
Speaker 1: things that we have traction on right now in the

708
00:37:59,760 --> 00:38:05,680
Speaker 1: market market are around business process, automation, digital labor, those

709
00:38:05,760 --> 00:38:08,960
Speaker 1: kind of things. But my other little piece of advice

710
00:38:09,000 --> 00:38:11,000
Speaker 1: there is keep it simple to begin with. You're going

711
00:38:11,080 --> 00:38:13,760
Speaker 1: to learn a lot, but getting started means you'll start

712
00:38:13,800 --> 00:38:17,560
Speaker 1: that learning curve. I even advise you my friends like, hey,

713
00:38:17,640 --> 00:38:19,359
Speaker 1: should I be playing around with some.

714
00:38:19,320 --> 00:38:21,640
Speaker 2: Of this AI stuff? And I say yeah, because I

715
00:38:21,680 --> 00:38:24,279
Speaker 2: think it will help you start to be more comfortable

716
00:38:24,680 --> 00:38:26,759
Speaker 2: and you may find a use case personally for that.

717
00:38:26,800 --> 00:38:29,600
Speaker 2: I think the same is true for businesses. The first

718
00:38:29,640 --> 00:38:33,440
Speaker 2: step in that journey is always with what data. Notice

719
00:38:33,440 --> 00:38:37,760
Speaker 2: when I talked about our customer support people. I thought about, Okay,

720
00:38:37,840 --> 00:38:40,920
Speaker 2: what's the data. The data is all of those logs

721
00:38:40,960 --> 00:38:44,040
Speaker 2: of all of those service engagements around the world, and

722
00:38:44,080 --> 00:38:45,879
Speaker 2: what could I do with that? Well, I could use

723
00:38:45,880 --> 00:38:48,520
Speaker 2: that to get to a knowledge base that really helps

724
00:38:49,360 --> 00:38:51,920
Speaker 2: and hopefully that I can do it in multiple languages

725
00:38:52,000 --> 00:38:54,480
Speaker 2: because it's global and I can you know, all of

726
00:38:54,520 --> 00:38:56,799
Speaker 2: those things. That was kind of my data sent That

727
00:38:56,840 --> 00:38:59,279
Speaker 2: one's not super simple, but we've had a lot of

728
00:38:59,320 --> 00:39:02,520
Speaker 2: experience in AI for other people that might just be

729
00:39:03,040 --> 00:39:06,600
Speaker 2: how do I automate filling out travel expense reports for

730
00:39:06,800 --> 00:39:09,719
Speaker 2: my company? We can help people that we have consulting,

731
00:39:09,760 --> 00:39:11,920
Speaker 2: we have wats and X tools. We can do that

732
00:39:12,280 --> 00:39:14,880
Speaker 2: like this, and we're doing it globally for people around

733
00:39:14,920 --> 00:39:17,239
Speaker 2: the world. Pick that thing. What's the data you have?

734
00:39:17,360 --> 00:39:20,920
Speaker 2: In that case, it's data of expense reports and it's like, okay,

735
00:39:20,960 --> 00:39:23,319
Speaker 2: we can help you automate that for people where they

736
00:39:23,320 --> 00:39:26,440
Speaker 2: could do it just by you know, a verbal interface.

737
00:39:26,880 --> 00:39:28,799
Speaker 2: What did you spend, where did you go, who you

738
00:39:28,840 --> 00:39:31,240
Speaker 2: were you with? Okay, we filled out your travel expense

739
00:39:31,239 --> 00:39:33,160
Speaker 2: report for you and you don't have to mess around

740
00:39:33,239 --> 00:39:33,480
Speaker 2: with it.

741
00:39:33,600 --> 00:39:36,160
Speaker 1: So we were playing with this idea where we would

742
00:39:36,560 --> 00:39:39,920
Speaker 1: pick a business and go in there and do it

743
00:39:39,960 --> 00:39:43,160
Speaker 1: would be AI makeover. Yeah, I love that. What's okay?

744
00:39:43,200 --> 00:39:46,759
Speaker 1: What is the ideal business to do? We only have

745
00:39:46,800 --> 00:39:49,080
Speaker 1: a couple months. We don't want to spend a kajillion dollars.

746
00:39:49,320 --> 00:39:52,000
Speaker 1: We want to be able to show tangibly and quickly

747
00:39:52,040 --> 00:39:55,000
Speaker 1: what AI can do. What's the ideal business to do

748
00:39:55,040 --> 00:39:56,920
Speaker 1: that in it can be a small business. We're not talking.

749
00:39:57,239 --> 00:39:58,600
Speaker 1: This isn't a grand corporate thing.

750
00:39:58,680 --> 00:40:03,799
Speaker 2: There boy small business that we could do and hey,

751
00:40:03,880 --> 00:40:08,120
Speaker 2: I make over. Customer support is one of my favorites

752
00:40:08,200 --> 00:40:11,040
Speaker 2: because it's a it's it's I have it on the

753
00:40:11,040 --> 00:40:14,400
Speaker 2: business side where I provide customer support. I have it

754
00:40:14,440 --> 00:40:17,399
Speaker 2: on the consumer side, where it drives me nuts when

755
00:40:17,400 --> 00:40:20,320
Speaker 2: I have to go through thirty layers of phone menus.

756
00:40:20,840 --> 00:40:22,880
Speaker 2: Speak to an agent, Speak to an agent, speak to

757
00:40:22,920 --> 00:40:27,120
Speaker 2: an agent. That for any business, I think is just

758
00:40:27,360 --> 00:40:29,600
Speaker 2: ripe to be able to kind of say, why do

759
00:40:29,680 --> 00:40:31,960
Speaker 2: I have to click through these manucent messages? I just

760
00:40:32,000 --> 00:40:34,600
Speaker 2: need to tell you in human language, here's the issue,

761
00:40:34,600 --> 00:40:37,360
Speaker 2: and I'll be really good about telling you details about

762
00:40:37,880 --> 00:40:40,320
Speaker 2: you know, I tried to set up this thing for

763
00:40:40,480 --> 00:40:42,239
Speaker 2: my bank and I do da da da da da.

764
00:40:42,520 --> 00:40:46,200
Speaker 2: They can go through all the menus automate that process.

765
00:40:46,480 --> 00:40:48,800
Speaker 2: I think it would change everything because all that frustration

766
00:40:48,960 --> 00:40:52,359
Speaker 2: as a consumer would go down dramatically. And it's all,

767
00:40:52,880 --> 00:40:55,880
Speaker 2: you know, why are you making me the beep booth

768
00:40:56,080 --> 00:41:00,920
Speaker 2: press one load press exactly, Well, don't offload me. Offload

769
00:41:00,920 --> 00:41:02,879
Speaker 2: to AI. We can help you with that.

770
00:41:03,160 --> 00:41:06,440
Speaker 1: Here's my version of that drives me crazy. Every morning

771
00:41:06,600 --> 00:41:09,640
Speaker 1: I go to the same coffee shop and I get

772
00:41:10,239 --> 00:41:13,640
Speaker 1: a cup of tea and a croissant. And here's what happens.

773
00:41:13,640 --> 00:41:16,359
Speaker 1: The person has their screen and they go I go,

774
00:41:16,640 --> 00:41:24,880
Speaker 1: cup of tea, croissant, sparkling water, like at least twenty keystrokes,

775
00:41:25,640 --> 00:41:28,279
Speaker 1: and then like then the screen is turning around. Like

776
00:41:28,360 --> 00:41:30,719
Speaker 1: at this point we're like forty five seconds in. I'm like,

777
00:41:31,000 --> 00:41:32,920
Speaker 1: why is this? First of all, it's not for me,

778
00:41:33,000 --> 00:41:36,920
Speaker 1: all those keystrokes, it's their internal right, So they're burdening

779
00:41:37,000 --> 00:41:37,960
Speaker 1: me in order to service.

780
00:41:38,000 --> 00:41:39,640
Speaker 2: To back it, you should be able to walk in,

781
00:41:39,960 --> 00:41:42,120
Speaker 2: go up and they go, I'm malcom the same thing,

782
00:41:42,760 --> 00:41:45,040
Speaker 2: and you just go yes, and then we're done.

783
00:41:45,120 --> 00:41:47,200
Speaker 1: Can we do AI makeover of my coffee shop?

784
00:41:49,120 --> 00:41:52,799
Speaker 2: You notice I quickly jumped more to banks than your

785
00:41:52,880 --> 00:41:55,759
Speaker 2: coffee shop because I think I'm a business person, but

786
00:41:56,160 --> 00:41:58,200
Speaker 2: I'm not trying to kind of do a deal on

787
00:41:58,320 --> 00:41:59,160
Speaker 2: one coffee shop.

788
00:41:59,480 --> 00:42:01,400
Speaker 1: But this isn't interesting because it takes me back to

789
00:42:01,480 --> 00:42:04,440
Speaker 1: something you said that I thought was really important. When

790
00:42:04,440 --> 00:42:07,200
Speaker 1: you were talking about when you were using AI and

791
00:42:07,280 --> 00:42:10,399
Speaker 1: your customer service thing, it was clear that your goal

792
00:42:10,600 --> 00:42:13,040
Speaker 1: you could have any number of goals, yes, going in.

793
00:42:13,320 --> 00:42:16,360
Speaker 1: It could be to cut costs, it could be to

794
00:42:16,480 --> 00:42:20,520
Speaker 1: dramatically improved profits. Your goal quite specifically, was to improve

795
00:42:20,600 --> 00:42:22,319
Speaker 1: the experience of your customer, right, so.

796
00:42:22,360 --> 00:42:24,680
Speaker 2: You were using it to that. All the other things

797
00:42:24,840 --> 00:42:27,600
Speaker 2: come from that come from. That is actually one of

798
00:42:27,680 --> 00:42:32,280
Speaker 2: the beautiful pillars of the IBM culture is delighting clients

799
00:42:32,480 --> 00:42:35,080
Speaker 2: is actually where all of the good stuff comes from.

800
00:42:35,239 --> 00:42:39,080
Speaker 1: So my coffee shop thing is the same principle. Right now,

801
00:42:39,520 --> 00:42:42,440
Speaker 1: they're making my customer experience worse and they don't want to,

802
00:42:43,040 --> 00:42:46,120
Speaker 1: but their eyes are glued to the space a moment

803
00:42:46,200 --> 00:42:47,960
Speaker 1: when I walk in and I want to say, Hi,

804
00:42:48,120 --> 00:42:51,359
Speaker 1: how are you doing? We could have a conversation. You're

805
00:42:51,360 --> 00:42:52,680
Speaker 1: too busy, busy baby pooping.

806
00:42:52,840 --> 00:42:54,480
Speaker 2: So like this is the same thing.

807
00:42:54,600 --> 00:42:56,360
Speaker 1: If they had it that oh this isn't if they

808
00:42:56,480 --> 00:42:59,080
Speaker 1: understood they had an opportunity to improve the experience of

809
00:42:59,160 --> 00:43:00,359
Speaker 1: their customer experience.

810
00:43:00,719 --> 00:43:04,560
Speaker 2: I would not be surprised if a chain comes along

811
00:43:05,040 --> 00:43:07,480
Speaker 2: where that is their value proposition. I would not be

812
00:43:07,560 --> 00:43:11,320
Speaker 2: surprised at all. Yeah, yeah, right, So I mean and

813
00:43:12,000 --> 00:43:15,240
Speaker 2: and when those things kind of catch hold, it becomes

814
00:43:15,280 --> 00:43:15,880
Speaker 2: a revolution.

815
00:43:16,160 --> 00:43:18,080
Speaker 1: You know, when the guy comes to do like to

816
00:43:18,160 --> 00:43:20,759
Speaker 1: redo your roof and they put a sign out front,

817
00:43:20,880 --> 00:43:23,440
Speaker 1: like you know, Joe's roofing. You guys could do the

818
00:43:23,480 --> 00:43:27,200
Speaker 1: same with my coffee shop. But like I'd be Ire

819
00:43:27,400 --> 00:43:27,960
Speaker 1: was here.

820
00:43:29,680 --> 00:43:30,800
Speaker 2: Exactly exactly.

821
00:43:32,760 --> 00:43:36,920
Speaker 1: In five years, the main frame will be dot dot

822
00:43:37,600 --> 00:43:42,799
Speaker 1: going strong, the mainframe going strong and.

823
00:43:43,200 --> 00:43:47,800
Speaker 2: With new capabilities, continuous new capabilities. I think when we

824
00:43:47,880 --> 00:43:51,800
Speaker 2: announced the last version, Z sixteen, the latest version, I

825
00:43:51,880 --> 00:43:55,719
Speaker 2: should say, and we said, hey, there's AI processing built

826
00:43:55,800 --> 00:43:58,279
Speaker 2: into it. This was before everybody was talking about that.

827
00:43:58,640 --> 00:44:00,759
Speaker 2: I think a lot of people thought, what's that for?

828
00:44:01,200 --> 00:44:04,080
Speaker 2: And we did it specifically for traditional AI fraud detection,

829
00:44:04,239 --> 00:44:07,360
Speaker 2: et cetera. This next version, not only do we have

830
00:44:07,440 --> 00:44:10,480
Speaker 2: the traditional AI built in, but we have optional cards

831
00:44:10,880 --> 00:44:12,960
Speaker 2: that you can plug into it to allow you to

832
00:44:13,040 --> 00:44:17,600
Speaker 2: do large language models for the enhanced fraud detection cases

833
00:44:17,640 --> 00:44:20,839
Speaker 2: that we talked about, where you know, it's more than

834
00:44:21,080 --> 00:44:24,680
Speaker 2: just what transactions were happening. So if you take that

835
00:44:24,840 --> 00:44:28,960
Speaker 2: and say, okay, the next generations, we have more transaction

836
00:44:29,120 --> 00:44:32,360
Speaker 2: volume than we've ever had in mainframes. Today, the business

837
00:44:32,520 --> 00:44:36,160
Speaker 2: is growing, it's strong, we keep innovating. In five years

838
00:44:36,200 --> 00:44:37,200
Speaker 2: it'll be going strong.

839
00:44:37,360 --> 00:44:39,920
Speaker 1: But we're people. You're saying this in the context of

840
00:44:40,840 --> 00:44:43,040
Speaker 1: for years people were predicting, weren't they that the main

841
00:44:43,080 --> 00:44:44,080
Speaker 1: brame was going to go away.

842
00:44:46,000 --> 00:44:48,399
Speaker 2: There were pundits in the market that said everything will

843
00:44:48,440 --> 00:44:50,400
Speaker 2: go away there, no one will ever have a box,

844
00:44:50,440 --> 00:44:53,080
Speaker 2: It'll all be online. I think this is something I've

845
00:44:53,160 --> 00:44:57,600
Speaker 2: learned big time in my long career. You know in

846
00:44:57,800 --> 00:45:01,560
Speaker 2: the IT industry is I don't believe everything you hear.

847
00:45:01,880 --> 00:45:05,920
Speaker 2: So I went back for my master's degree at Stanford

848
00:45:05,960 --> 00:45:10,760
Speaker 2: after I had worked a while in as a hardware designer,

849
00:45:11,080 --> 00:45:13,880
Speaker 2: and everybody told me be sure to do your masters

850
00:45:13,920 --> 00:45:16,919
Speaker 2: in software. Hardware is dead. I went on to work

851
00:45:17,200 --> 00:45:20,720
Speaker 2: for thirty plus years in hardware and infrastructure. Now software

852
00:45:20,760 --> 00:45:23,160
Speaker 2: became important, and I'm glad I had that extra training

853
00:45:23,239 --> 00:45:26,000
Speaker 2: in software because it helped me in hardware. But hardware

854
00:45:26,080 --> 00:45:29,640
Speaker 2: wasn't dead. Then I heard all infrastructure will go into

855
00:45:29,680 --> 00:45:32,640
Speaker 2: the cloud. There won't be that hasn't happened. It's not happening.

856
00:45:33,000 --> 00:45:35,400
Speaker 2: Then I heard there will only be one cloud because

857
00:45:35,480 --> 00:45:37,920
Speaker 2: one of the players will dominate. There's not one cloud.

858
00:45:38,000 --> 00:45:42,239
Speaker 2: So I think it's as humans we like to oversimplify

859
00:45:42,239 --> 00:45:44,440
Speaker 2: and go, oh, it's all going to be this, And

860
00:45:44,640 --> 00:45:48,200
Speaker 2: kind of what I've learned is fit for purpose matters

861
00:45:48,760 --> 00:45:54,440
Speaker 2: in everything. It matters in size of infrastructure, it matters

862
00:45:54,520 --> 00:45:57,160
Speaker 2: in the stack that goes along with solving a specific

863
00:45:57,400 --> 00:46:00,480
Speaker 2: use case. If you're willing to design something that's the

864
00:46:00,640 --> 00:46:02,960
Speaker 2: best at that use case, if you're willing to design

865
00:46:03,000 --> 00:46:05,279
Speaker 2: the coffee shop that is the best at greeting me,

866
00:46:05,680 --> 00:46:07,520
Speaker 2: there's a spot for you, and there may be a

867
00:46:07,600 --> 00:46:11,680
Speaker 2: big business in doing that. So oversimplifying is really when you.

868
00:46:11,760 --> 00:46:15,240
Speaker 1: Heard all those predictions, did you believe them at the time.

869
00:46:16,880 --> 00:46:19,600
Speaker 2: They looked like they were trending in that direction. I'll

870
00:46:19,640 --> 00:46:22,279
Speaker 2: tell you some right now which might be useful. There

871
00:46:22,320 --> 00:46:24,719
Speaker 2: will only be one GPU company and they're going to

872
00:46:25,440 --> 00:46:28,000
Speaker 2: end up taking over the world. It's a pretty obvious answer.

873
00:46:28,080 --> 00:46:31,440
Speaker 2: Whose economic values risen dramatically. I don't think that's going

874
00:46:31,520 --> 00:46:33,920
Speaker 2: to be the case. In fact, I think that ninety

875
00:46:34,000 --> 00:46:39,560
Speaker 2: percent of processing for AI actually happen happens at inferencing,

876
00:46:40,080 --> 00:46:43,560
Speaker 2: and inferencing is not as GPU and hardware intensive as

877
00:46:43,600 --> 00:46:46,000
Speaker 2: the other things and is a lot more amenable to

878
00:46:46,160 --> 00:46:49,200
Speaker 2: fit for purpose. So the model size will matter. The

879
00:46:49,360 --> 00:46:51,640
Speaker 2: tuning matters a lot. As we're learning. We have a

880
00:46:51,719 --> 00:46:55,759
Speaker 2: product around instruct lab that's really focused on tuning. So

881
00:46:56,200 --> 00:46:58,120
Speaker 2: that was one thing is there'll be one GPU. The

882
00:46:58,200 --> 00:47:02,080
Speaker 2: other thing is that the biggest model will win, I

883
00:47:02,160 --> 00:47:04,520
Speaker 2: think is another thing that's kind of people are saying

884
00:47:04,600 --> 00:47:06,640
Speaker 2: right now. Don't believe that I believe they will be

885
00:47:06,800 --> 00:47:10,040
Speaker 2: fit for purpose models. It takes a lot of money

886
00:47:10,080 --> 00:47:13,360
Speaker 2: to run to create a huge model, and then to

887
00:47:13,560 --> 00:47:16,279
Speaker 2: run a huge model, or to even infer off of

888
00:47:16,320 --> 00:47:19,560
Speaker 2: a huge model. I don't need a massive training GPU

889
00:47:19,840 --> 00:47:23,800
Speaker 2: set thing to solve my thirteen thousand people customer support issues.

890
00:47:23,840 --> 00:47:25,840
Speaker 2: So why would I feel like I got to go

891
00:47:26,080 --> 00:47:28,440
Speaker 2: farm that out for a big expensive thing. I can

892
00:47:28,520 --> 00:47:30,480
Speaker 2: do that on a small box. In some cases I

893
00:47:30,560 --> 00:47:32,240
Speaker 2: might even be able to do that on a laptop.

894
00:47:32,760 --> 00:47:34,600
Speaker 2: The other thing I'll say in this we are so

895
00:47:34,880 --> 00:47:37,560
Speaker 2: early innings in AI. A lot of things are going

896
00:47:37,640 --> 00:47:40,000
Speaker 2: to change. So anybody kind of saying it will all

897
00:47:40,080 --> 00:47:42,440
Speaker 2: be X, Y or Z, I just think you have

898
00:47:42,640 --> 00:47:44,560
Speaker 2: no idea how this is going to play out, and

899
00:47:45,239 --> 00:47:46,880
Speaker 2: it's up to us to go figure out how it

900
00:47:46,920 --> 00:47:47,399
Speaker 2: plays out.

901
00:47:47,719 --> 00:47:51,600
Speaker 1: Yeah, yeah, all right, in five years, AI will be

902
00:47:51,960 --> 00:47:54,000
Speaker 1: dot dot dot still new.

903
00:47:56,600 --> 00:48:00,560
Speaker 2: It will have moved a bunch in five years, but

904
00:48:00,719 --> 00:48:05,040
Speaker 2: the potential for the disruption in the world will still

905
00:48:05,160 --> 00:48:08,120
Speaker 2: will still be very early innings in that process. And

906
00:48:08,200 --> 00:48:10,520
Speaker 2: I think that's super important to realize. That's why I

907
00:48:10,600 --> 00:48:14,000
Speaker 2: say get started, start thinking about how that could change,

908
00:48:14,000 --> 00:48:16,799
Speaker 2: because it'll be some little things first, but it will

909
00:48:16,880 --> 00:48:18,440
Speaker 2: continue to snowball. This is.

910
00:48:20,000 --> 00:48:24,520
Speaker 1: A common observation that we the invention of the capability

911
00:48:26,000 --> 00:48:28,400
Speaker 1: massively predates the understanding of.

912
00:48:28,440 --> 00:48:30,640
Speaker 2: The capability, right, Like I love that.

913
00:48:31,000 --> 00:48:36,799
Speaker 1: Yeah, Like yes, recorded recording shows on television is invented

914
00:48:36,920 --> 00:48:42,560
Speaker 1: in the sixties, probably the VCR. We don't really understand

915
00:48:42,719 --> 00:48:47,279
Speaker 1: what it's used for until the oughts. Was what was

916
00:48:47,320 --> 00:48:50,719
Speaker 1: really good for is being able to tell a story sequentially, Yes,

917
00:48:51,080 --> 00:48:53,280
Speaker 1: over time, because you know that the person will always

918
00:48:53,280 --> 00:48:55,400
Speaker 1: see in the episode before, so you got the Sopranos,

919
00:48:55,480 --> 00:48:59,279
Speaker 1: And yes, yes, Hollywood wanted to ban the VCR in

920
00:48:59,320 --> 00:49:01,799
Speaker 1: the beginning. Yeah, because they thought it was good. They

921
00:49:01,840 --> 00:49:04,920
Speaker 1: thought the point of it was thought THEYDN understand No, No, no,

922
00:49:05,080 --> 00:49:08,480
Speaker 1: it's storytelling. It's actually your business is getting better. Yes, yes,

923
00:49:08,760 --> 00:49:10,480
Speaker 1: took them twenty years to figure that out, which is

924
00:49:10,760 --> 00:49:13,080
Speaker 1: to your point, why would we know what AI was

925
00:49:13,120 --> 00:49:13,839
Speaker 1: four and five years.

926
00:49:14,080 --> 00:49:16,160
Speaker 2: Well, that's why you hear people kind of say, oh

927
00:49:16,239 --> 00:49:18,800
Speaker 2: my gosh, AI, that's that will just eliminate jobs. No,

928
00:49:19,000 --> 00:49:21,000
Speaker 2: it'll make jobs better. That's how I view it.

929
00:49:21,200 --> 00:49:24,640
Speaker 1: Yeah, what's the number one thing that people misunderstand about AI?

930
00:49:24,840 --> 00:49:27,960
Speaker 2: Is that it that it'll I think that's that. That

931
00:49:28,120 --> 00:49:30,960
Speaker 2: would be the human kind of understanding part of it.

932
00:49:31,120 --> 00:49:34,400
Speaker 2: The technology part of it, I think would be what

933
00:49:34,560 --> 00:49:38,719
Speaker 2: I was talking about, fit for purpose, meaning that it

934
00:49:38,840 --> 00:49:41,600
Speaker 2: isn't just going to be a GPU arms race all

935
00:49:41,680 --> 00:49:44,239
Speaker 2: of AI. I don't believe that at all. It will

936
00:49:44,320 --> 00:49:45,920
Speaker 2: change everything, but it's not just going to be a

937
00:49:46,000 --> 00:49:49,840
Speaker 2: GPU armed Next question, what advice would you give yourself

938
00:49:49,920 --> 00:49:52,000
Speaker 2: ten years ago to better prepare you for today?

939
00:49:52,480 --> 00:49:57,400
Speaker 1: I'm changing this question, okay. I would say, let's imagine

940
00:49:57,480 --> 00:50:00,800
Speaker 1: that what was your what what college you to go to?

941
00:50:01,680 --> 00:50:04,920
Speaker 2: I went to three of them. My undergrad was Utah

942
00:50:05,000 --> 00:50:08,320
Speaker 2: State University, my NBA was Santa Clara University, and my

943
00:50:08,640 --> 00:50:10,560
Speaker 2: master's in w was Stanford. OK.

944
00:50:11,920 --> 00:50:13,600
Speaker 1: Any one of those three called you up and says

945
00:50:14,000 --> 00:50:18,840
Speaker 1: we want you to give the commencement address, and imagine

946
00:50:18,880 --> 00:50:22,279
Speaker 1: it's let's just safe the sake of argument, it's just

947
00:50:22,360 --> 00:50:25,239
Speaker 1: to the stamp people, because those are the relevant parties here.

948
00:50:26,280 --> 00:50:29,160
Speaker 2: What do you tell them? Boy? What do I tell them?

949
00:50:33,040 --> 00:50:37,480
Speaker 2: Let's see, I think I would start with life is

950
00:50:37,520 --> 00:50:40,600
Speaker 2: a marathon, not a sprint. It would be the first one.

951
00:50:42,320 --> 00:50:44,800
Speaker 2: The second thing I would say that in that spirit

952
00:50:45,200 --> 00:50:48,960
Speaker 2: is be sure to set yourself some big, hairy, audacious

953
00:50:49,000 --> 00:50:54,040
Speaker 2: goals and don't be overly disappointed if you don't hit

954
00:50:54,160 --> 00:50:59,319
Speaker 2: them all. Going after those big hairy, audacious goals will

955
00:50:59,360 --> 00:51:02,399
Speaker 2: get you on a path where you will learn so much.

956
00:51:03,040 --> 00:51:06,160
Speaker 2: You will achieve more than you ever could imagine you

957
00:51:06,200 --> 00:51:08,600
Speaker 2: would have achieved. That's what the advice I give to

958
00:51:08,680 --> 00:51:11,400
Speaker 2: my kids is, Yeah, set some big goals, get after it.

959
00:51:12,120 --> 00:51:13,759
Speaker 2: You may or may not achieve them, but you'll be

960
00:51:13,840 --> 00:51:15,440
Speaker 2: better for the whole process when you're done.

961
00:51:15,440 --> 00:51:17,759
Speaker 1: By the way, as someone whose kids are younger than yours,

962
00:51:18,640 --> 00:51:20,880
Speaker 1: is it actually useful to give your give advice to

963
00:51:20,920 --> 00:51:23,680
Speaker 1: your kids, the pointments exercise TVD.

964
00:51:23,920 --> 00:51:25,799
Speaker 2: We're still on the journey and I think we will

965
00:51:25,880 --> 00:51:27,680
Speaker 2: be for a long time.

966
00:51:27,880 --> 00:51:30,640
Speaker 1: I don't know how are you already using AI in

967
00:51:30,760 --> 00:51:31,839
Speaker 1: your day to day life today?

968
00:51:33,960 --> 00:51:38,080
Speaker 2: Personally, I would say it's replacing a good chunk of

969
00:51:38,160 --> 00:51:41,120
Speaker 2: my search. You know, I'm less likely to go blindly

970
00:51:41,480 --> 00:51:44,240
Speaker 2: stumbling through a bunch of web pages looking for stuff.

971
00:51:44,680 --> 00:51:47,000
Speaker 2: I'm more likely to ask a question from a few

972
00:51:47,239 --> 00:51:49,960
Speaker 2: AI engines kind of see, get me in the right direction,

973
00:51:50,120 --> 00:51:52,680
Speaker 2: then I'll go bumble through a few things. At work,

974
00:51:53,200 --> 00:51:58,560
Speaker 2: I can tell you code development right now, we are

975
00:51:58,600 --> 00:52:02,600
Speaker 2: seeing massive improvements and code development and support products We

976
00:52:02,760 --> 00:52:07,520
Speaker 2: have like Watson Code Assistant that is really showing immediate

977
00:52:07,640 --> 00:52:10,680
Speaker 2: return for a code developers, and I think that will

978
00:52:10,840 --> 00:52:14,680
Speaker 2: again be a tool that increases productivity for code developers

979
00:52:15,120 --> 00:52:16,960
Speaker 2: immediately across the globe. Yeah.

980
00:52:17,800 --> 00:52:21,120
Speaker 1: Last question, what's the one skill that every technology leader

981
00:52:21,320 --> 00:52:23,480
Speaker 1: needs that has nothing to do with technology.

982
00:52:24,480 --> 00:52:28,560
Speaker 2: Being able to inspire a set of people toward a

983
00:52:28,640 --> 00:52:32,080
Speaker 2: common goal and collaborate to achieve it. That's at the

984
00:52:32,200 --> 00:52:35,960
Speaker 2: core of everything everything. That's a lovely way to end.

985
00:52:36,640 --> 00:52:37,920
Speaker 2: Thank you so much, Rick, Thank you.

986
00:52:40,480 --> 00:52:44,760
Speaker 1: This conversation left me excited. I'm now imagining the potential

987
00:52:44,840 --> 00:52:47,560
Speaker 1: for new use cases for AI in all sorts of

988
00:52:47,680 --> 00:52:51,400
Speaker 1: different businesses. Rick didn't seem soldom my idea of a

989
00:52:51,400 --> 00:52:54,880
Speaker 1: coffee shop makeover, but it's clear there's lots of opportunities

990
00:52:54,920 --> 00:52:58,759
Speaker 1: here to increase speed and efficiency, to achieve your objectives,

991
00:52:59,000 --> 00:53:02,480
Speaker 1: and to dream beyond the current applications for this technology.

992
00:53:03,480 --> 00:53:05,720
Speaker 1: At the end of the day, the scaling of AI

993
00:53:06,120 --> 00:53:09,440
Speaker 1: will rely on the right infrastructure to support it. With

994
00:53:09,560 --> 00:53:12,680
Speaker 1: the right tools, you can solve problems that are unique

995
00:53:12,719 --> 00:53:27,520
Speaker 1: tier industry and improve the experience for your customers. Smart

996
00:53:27,560 --> 00:53:30,840
Speaker 1: Talks with IBM is produced by Matt Romano, Amy Gains,

997
00:53:30,920 --> 00:53:35,160
Speaker 1: McQuaid and Jacob Goldstein. Were edited by Lydia gene Kott,

998
00:53:35,640 --> 00:53:41,200
Speaker 1: mastering by Jake Korsky. Theme song by Gramoscope. Special thanks

999
00:53:41,239 --> 00:53:43,759
Speaker 1: to the eight Bar and IBM teams, as well as

1000
00:53:43,800 --> 00:53:47,680
Speaker 1: the Pushkin Marketing team. Smart Talks with IBM is a

1001
00:53:47,719 --> 00:53:54,040
Speaker 1: production of Pushkin Industries and Ruby Studio at iHeartMedia. To

1002
00:53:54,120 --> 00:53:59,200
Speaker 1: find more Pushkin podcasts, listen on the iHeartRadio app, Apple Podcasts,

1003
00:53:59,600 --> 00:54:03,760
Speaker 1: or where ever you listen to podcasts. I'm Malcolm Gladwell.

1004
00:54:08,800 --> 00:54:12,520
Speaker 1: This is a paid advertisement from IBM. The conversations on

1005
00:54:12,600 --> 00:54:18,759
Speaker 1: this podcast don't necessarily represent IBM's positions, strategies, or opinions.