1
00:00:00,040 --> 00:00:02,800
Speaker 1: Hey everyone, it's Robert and Joe here. Today we've got

2
00:00:02,840 --> 00:00:05,080
Speaker 1: something a little different to share with you. It's a

3
00:00:05,080 --> 00:00:08,719
Speaker 1: new season of the Smart Talks with IBM podcast series.

4
00:00:09,160 --> 00:00:11,959
Speaker 2: This season, on smart Talks, Malcolm Gladwell and team are

5
00:00:12,039 --> 00:00:15,240
Speaker 2: diving into the transformative world of artificial intelligence with a

6
00:00:15,240 --> 00:00:18,560
Speaker 2: fresh perspective on the concept of open What does open

7
00:00:18,640 --> 00:00:21,840
Speaker 2: really mean in the context of AI. It can mean

8
00:00:21,960 --> 00:00:25,600
Speaker 2: open source code or open data, but it also encompasses

9
00:00:25,680 --> 00:00:30,760
Speaker 2: fostering an ecosystem of ideas, ensuring diverse perspectives are heard,

10
00:00:31,120 --> 00:00:33,479
Speaker 2: and enabling new levels of transparency.

11
00:00:33,800 --> 00:00:37,040
Speaker 1: Join hosts from your favorite pushkin podcasts as they explore

12
00:00:37,080 --> 00:00:40,879
Speaker 1: how opennes in AI is reshaping industries, driving innovation, and

13
00:00:40,960 --> 00:00:44,800
Speaker 1: redefining what's possible. You'll hear from industry experts and leaders

14
00:00:44,840 --> 00:00:48,320
Speaker 1: about the implication and possibilities of open AI, and of course,

15
00:00:48,640 --> 00:00:50,879
Speaker 1: Malcolm Gladwell will be there to guide you through the

16
00:00:50,920 --> 00:00:52,640
Speaker 1: season with his unique insights.

17
00:00:52,960 --> 00:00:55,520
Speaker 2: Look out for new episodes of Smart Talks every other

18
00:00:55,560 --> 00:00:59,160
Speaker 2: week on the iHeartRadio app, Apple Podcasts, or wherever you

19
00:00:59,200 --> 00:01:02,480
Speaker 2: get your podcasts, and learn more at IBM dot com

20
00:01:02,480 --> 00:01:03,840
Speaker 2: slash smart Talks.

21
00:01:05,560 --> 00:01:08,959
Speaker 3: Hello, Hello, Welcome to smart Talks with IBM, a podcast

22
00:01:09,240 --> 00:01:14,880
Speaker 3: from Pushkin Industries, iHeartRadio and IBM. I'm Malcolm Godwell. This season,

23
00:01:14,920 --> 00:01:17,920
Speaker 3: we're diving back into the world of artificial intelligence, but

24
00:01:17,959 --> 00:01:23,840
Speaker 3: with a focus on the powerful concept of open its possibilities, implications,

25
00:01:23,880 --> 00:01:28,559
Speaker 3: and misconceptions. On today's episode, our season finale, I'm joined

26
00:01:28,560 --> 00:01:32,840
Speaker 3: by Rick Lewis, the senior vice president of Infrastructure at IBM.

27
00:01:33,440 --> 00:01:36,920
Speaker 3: Rick has had a remarkable career focused around product innovation.

28
00:01:37,640 --> 00:01:40,600
Speaker 3: He was actually a few years into retirement when IBM

29
00:01:40,680 --> 00:01:46,080
Speaker 3: came calling with an opportunity he just couldn't turn down. Thankfully,

30
00:01:46,280 --> 00:01:49,600
Speaker 3: Rick came out of retirement and today he oversees a

31
00:01:49,680 --> 00:01:55,160
Speaker 3: vast portfolio from storage and software to global customer support operations,

32
00:01:55,560 --> 00:01:57,880
Speaker 3: and he's engaged in one of the key problems facing

33
00:01:57,920 --> 00:02:02,800
Speaker 3: companies today, an explosion of data. In talking with Rick,

34
00:02:02,880 --> 00:02:05,240
Speaker 3: I can see that this problem of having so much

35
00:02:05,360 --> 00:02:09,440
Speaker 3: data is also an incredible opportunity because if you're able

36
00:02:09,480 --> 00:02:12,160
Speaker 3: to leverage that data to get the most value out

37
00:02:12,160 --> 00:02:14,560
Speaker 3: of it, then you can use it to help bring

38
00:02:14,600 --> 00:02:18,360
Speaker 3: your business into the future. We talked about the serious

39
00:02:18,400 --> 00:02:21,280
Speaker 3: computing power needed to scale AI, as well as the

40
00:02:21,280 --> 00:02:25,480
Speaker 3: ways that infrastructure storage solutions can be essential to enabling

41
00:02:25,560 --> 00:02:30,079
Speaker 3: this new world of possibilities. It's a really great conversation,

42
00:02:30,600 --> 00:02:37,800
Speaker 3: so let's get to it. I'm here with Rick Lewis. Rick, Welcome,

43
00:02:38,240 --> 00:02:41,880
Speaker 3: Thank Here. We are in the IBM's New York City

44
00:02:41,880 --> 00:02:45,240
Speaker 3: headquarters at one Madison Avenue. I'm going to start with

45
00:02:46,040 --> 00:02:47,040
Speaker 3: you're a hardware guy.

46
00:02:47,240 --> 00:02:50,840
Speaker 4: I'm a hardware guy. I grew up doing hardware chip engineering.

47
00:02:51,280 --> 00:02:53,720
Speaker 4: But like I tell a lot of people, a chip

48
00:02:53,800 --> 00:02:57,360
Speaker 4: engineering project is actually a giant software project with a

49
00:02:57,480 --> 00:03:00,160
Speaker 4: piece of hardware at the end of the project. I

50
00:03:00,160 --> 00:03:02,919
Speaker 4: think if you have that analytical brain, you like to

51
00:03:03,000 --> 00:03:06,640
Speaker 4: solve problems, you'd like to get things working. You can

52
00:03:06,680 --> 00:03:07,960
Speaker 4: do that in soetwork.

53
00:03:08,040 --> 00:03:11,400
Speaker 3: But as being someone coming from a hardware background mean

54
00:03:11,440 --> 00:03:14,040
Speaker 3: that you think about problems in a different way.

55
00:03:14,680 --> 00:03:18,320
Speaker 4: I think one thing that you do from a hardware background,

56
00:03:18,320 --> 00:03:21,520
Speaker 4: and especially a chip background, is a chip spin and

57
00:03:21,560 --> 00:03:25,440
Speaker 4: costs millions of dollars, So you're a lot more likely

58
00:03:25,560 --> 00:03:28,639
Speaker 4: to make sure everything has a great chance of being

59
00:03:28,680 --> 00:03:31,160
Speaker 4: perfect from the get go. Or if you start kind

60
00:03:31,160 --> 00:03:34,200
Speaker 4: of from a software background, your general mindset is I

61
00:03:34,200 --> 00:03:35,840
Speaker 4: don't know, try this, see if it works. I don't know,

62
00:03:35,880 --> 00:03:37,440
Speaker 4: try that is if it work and you kind of

63
00:03:37,480 --> 00:03:40,600
Speaker 4: it it iterate chip people are a little more uptight

64
00:03:40,680 --> 00:03:44,200
Speaker 4: about okay, if this first round of the chip breaks

65
00:03:44,440 --> 00:03:47,119
Speaker 4: costs us from building another new round of the chip.

66
00:03:47,280 --> 00:03:50,360
Speaker 3: Yeah, so you're a little more You guys are spend

67
00:03:50,400 --> 00:03:51,160
Speaker 3: more time.

68
00:03:52,000 --> 00:03:56,240
Speaker 4: Planning and planning verifying, tons of time verifying.

69
00:03:57,400 --> 00:04:01,360
Speaker 3: So you began your career as you look back, yes, correct,

70
00:04:01,400 --> 00:04:02,520
Speaker 3: and you were there for how many years?

71
00:04:02,600 --> 00:04:03,920
Speaker 4: I was there for thirty two years?

72
00:04:04,160 --> 00:04:07,920
Speaker 3: Yes, And your last job there was I was leading.

73
00:04:07,600 --> 00:04:10,280
Speaker 4: The software defining cloud business. I had grown up a

74
00:04:10,320 --> 00:04:15,800
Speaker 4: hardware guy. I had done all kinds of hardware projects,

75
00:04:15,840 --> 00:04:20,360
Speaker 4: big complicated Unix servers and things like that, and then came,

76
00:04:20,720 --> 00:04:22,520
Speaker 4: you know, grew out of R and D and more

77
00:04:22,520 --> 00:04:27,279
Speaker 4: into the business realm. And then I'm much an innovator

78
00:04:27,320 --> 00:04:31,760
Speaker 4: at heart. I really like innovating new concepts things like that.

79
00:04:32,040 --> 00:04:35,240
Speaker 4: And what I learned is I enjoyed innovating business models

80
00:04:35,279 --> 00:04:38,400
Speaker 4: and software projects as much as I did hardware products

81
00:04:39,200 --> 00:04:42,920
Speaker 4: and projects, and so getting teams inspired towards doing that

82
00:04:43,320 --> 00:04:45,680
Speaker 4: was really a deep fascination for me. So I ended

83
00:04:45,760 --> 00:04:49,480
Speaker 4: up doing a fantastic variety of experiences and had a

84
00:04:49,520 --> 00:04:54,200
Speaker 4: successful run and honestly retired, intending to retire and do

85
00:04:54,279 --> 00:04:56,440
Speaker 4: some of my outside activities and things like that.

86
00:04:56,960 --> 00:04:59,360
Speaker 3: And then how long did you stay retired before IBM

87
00:04:59,400 --> 00:04:59,839
Speaker 3: can close?

88
00:05:00,120 --> 00:05:04,560
Speaker 4: Almost two years? And when I first got at ALL,

89
00:05:04,800 --> 00:05:07,800
Speaker 4: I thought, no, I'm having too much fun. But I

90
00:05:07,839 --> 00:05:12,719
Speaker 4: would say three things really got me thinking hard about it. One,

91
00:05:13,960 --> 00:05:16,440
Speaker 4: the industry that we're in, the IT industry. I think

92
00:05:16,440 --> 00:05:18,400
Speaker 4: it's the golden age. And what I mean by that

93
00:05:18,600 --> 00:05:22,240
Speaker 4: is for twenty years of that career, it is kind

94
00:05:22,240 --> 00:05:24,640
Speaker 4: of in the back office. Hey, make sure that stuff

95
00:05:24,680 --> 00:05:27,719
Speaker 4: doesn't crash, and can you please reduce the cost as

96
00:05:27,800 --> 00:05:29,800
Speaker 4: much as possible, because it's not that important to the

97
00:05:29,880 --> 00:05:33,320
Speaker 4: main business. It's just a back office function. You can

98
00:05:33,360 --> 00:05:35,599
Speaker 4: see it right now. It is at the forefront of

99
00:05:35,680 --> 00:05:40,200
Speaker 4: all business revolution. It happened with the Internet. It happened

100
00:05:40,200 --> 00:05:44,159
Speaker 4: again with cloud and how that changed every ounce of business,

101
00:05:44,240 --> 00:05:46,920
Speaker 4: not just IT business, but all business. And I think

102
00:05:46,920 --> 00:05:50,159
Speaker 4: it's happening again with AI. So to be in that

103
00:05:50,240 --> 00:05:52,719
Speaker 4: career that long and to miss the kind of this

104
00:05:52,839 --> 00:05:55,159
Speaker 4: age where it's like this is front and center. This

105
00:05:55,320 --> 00:05:59,560
Speaker 4: changes everything about all businesses, not just technology businesses. I

106
00:05:59,640 --> 00:06:03,279
Speaker 4: was kind of feeling like, gosh, you you trained to

107
00:06:03,360 --> 00:06:08,400
Speaker 4: be in these really awesome environments. Why wouldn't you do

108
00:06:08,480 --> 00:06:10,719
Speaker 4: that for a little while longer while you still can

109
00:06:10,800 --> 00:06:14,440
Speaker 4: do it. That combined with IBM and IBM seeing the

110
00:06:14,520 --> 00:06:18,719
Speaker 4: talent pool, the brilliant people at IBM, I worked with

111
00:06:18,760 --> 00:06:21,440
Speaker 4: a ton of brilliant people before I saw a chance

112
00:06:21,480 --> 00:06:24,080
Speaker 4: to work with even a larger staff of brilliant people.

113
00:06:24,200 --> 00:06:27,400
Speaker 4: And then the assets that IBM had, which is, you know,

114
00:06:27,440 --> 00:06:30,160
Speaker 4: they'd already been doing a lot of experimentation in AI,

115
00:06:30,279 --> 00:06:35,480
Speaker 4: they're working in quantum, the deep, rich heritage of successful projects.

116
00:06:35,480 --> 00:06:37,520
Speaker 4: I thought, who wouldn't want to kind of see if

117
00:06:37,520 --> 00:06:40,640
Speaker 4: they could be part of that next great wave of IBM.

118
00:06:40,760 --> 00:06:43,279
Speaker 4: And so I kind of decided, all right, I'm going

119
00:06:43,320 --> 00:06:45,360
Speaker 4: to put the outside interest on hold for a while

120
00:06:45,360 --> 00:06:46,400
Speaker 4: and get back in the game.

121
00:06:46,520 --> 00:06:49,839
Speaker 3: Along between the phone call, the first phone call and

122
00:06:49,920 --> 00:06:50,880
Speaker 3: you say, yes.

123
00:06:50,680 --> 00:06:54,920
Speaker 4: It was a while, It was probably six months. Arvin's

124
00:06:55,000 --> 00:06:57,600
Speaker 4: our CEO, teases me about that a lot. Yeah, he

125
00:06:57,720 --> 00:06:58,080
Speaker 4: was like, I.

126
00:06:58,040 --> 00:07:00,240
Speaker 3: Don't think six months is that long? It took a while.

127
00:07:00,440 --> 00:07:02,800
Speaker 3: You're a retirement I know. Yeah.

128
00:07:02,440 --> 00:07:05,680
Speaker 4: It's one thing to compare I'm working here and doing

129
00:07:05,680 --> 00:07:08,520
Speaker 4: this stuff versus working there. It's really hard to compare.

130
00:07:08,720 --> 00:07:11,240
Speaker 4: I'm doing exactly as I want to do every single

131
00:07:11,320 --> 00:07:13,160
Speaker 4: day when I wake up, and now I'm not going

132
00:07:13,200 --> 00:07:15,960
Speaker 4: to get to do that again. It took a while

133
00:07:15,960 --> 00:07:18,280
Speaker 4: for me to get over and I thought, I can't

134
00:07:18,320 --> 00:07:22,640
Speaker 4: miss this wave, and I'm really really happy that I did,

135
00:07:22,760 --> 00:07:26,440
Speaker 4: because we're doing some amazing, fun things and I'm getting

136
00:07:26,480 --> 00:07:28,960
Speaker 4: challenged in ways that I never did, so it's really fun.

137
00:07:29,200 --> 00:07:32,240
Speaker 3: Talk a little bit about your job here at IBM.

138
00:07:32,440 --> 00:07:36,600
Speaker 3: You oversee a kind of massive portfolio.

139
00:07:36,720 --> 00:07:39,640
Speaker 4: It's a big group, so I run the Infrastructure organization.

140
00:07:39,680 --> 00:07:43,000
Speaker 4: There's three main groups of products at IBM. There's the

141
00:07:43,080 --> 00:07:46,800
Speaker 4: Infrastructure group, which I run, the Software group, and the

142
00:07:46,840 --> 00:07:52,040
Speaker 4: Consulting group. And infrastructure is built up of mainframes, which

143
00:07:52,080 --> 00:07:56,240
Speaker 4: is called our Z portfolio, our servers which is our

144
00:07:56,280 --> 00:08:00,600
Speaker 4: power portfolio storage. By the way, those business include the

145
00:08:00,640 --> 00:08:02,960
Speaker 4: supply chain to build all of that stuff, so that's

146
00:08:02,960 --> 00:08:06,120
Speaker 4: in the group. Then I have the worldwide Customer Support

147
00:08:06,240 --> 00:08:10,520
Speaker 4: Organization it's called TLS Technology life Cycle Services, which is

148
00:08:10,560 --> 00:08:13,920
Speaker 4: a network of about thirteen thousand people around the globe

149
00:08:13,920 --> 00:08:16,040
Speaker 4: that make sure that everything runs and works when you

150
00:08:16,040 --> 00:08:19,760
Speaker 4: buy IBM products. And then also our IBM Cloud, which

151
00:08:19,800 --> 00:08:24,680
Speaker 4: is how we host applications and deliver as a service

152
00:08:24,720 --> 00:08:27,720
Speaker 4: products for our client base. So there's a lot. I

153
00:08:27,760 --> 00:08:30,880
Speaker 4: think it's about forty five thousand people total. Do those.

154
00:08:32,160 --> 00:08:34,959
Speaker 3: Components of the Infrastructure group are they aligned in their

155
00:08:35,000 --> 00:08:38,160
Speaker 3: trajectory or are they on different paths?

156
00:08:38,200 --> 00:08:41,080
Speaker 4: And I'm just curious what son little of both. It's

157
00:08:41,120 --> 00:08:44,000
Speaker 4: interesting you would ask that because I think of all

158
00:08:44,080 --> 00:08:48,079
Speaker 4: of the challenges coming to the new company, there were

159
00:08:48,080 --> 00:08:52,120
Speaker 4: things I expected, things that they didn't expect. But getting

160
00:08:52,200 --> 00:08:57,120
Speaker 4: that culture right in that group has been a big challenge.

161
00:08:57,800 --> 00:09:04,160
Speaker 4: IBM has a great culture toward quality products, toward emphasizing

162
00:09:04,240 --> 00:09:06,800
Speaker 4: passion for the client and making sure that the client

163
00:09:06,920 --> 00:09:10,400
Speaker 4: is happy, and for delivering innovation on a scale that

164
00:09:11,200 --> 00:09:13,880
Speaker 4: you know, for more than one hundred years has been

165
00:09:14,720 --> 00:09:18,760
Speaker 4: extremely powerful. But with success comes some challenges. And with

166
00:09:18,840 --> 00:09:22,959
Speaker 4: that success you can tend to get a little bit insular,

167
00:09:23,160 --> 00:09:25,360
Speaker 4: like you don't keep an eye on the competition. As well,

168
00:09:25,440 --> 00:09:27,960
Speaker 4: you can get more siloed, where you know, this is

169
00:09:28,040 --> 00:09:31,000
Speaker 4: my business unit, this is my business unit, I compete

170
00:09:31,000 --> 00:09:33,400
Speaker 4: with the other business unit. That's not a good thing

171
00:09:33,480 --> 00:09:37,920
Speaker 4: when you're a company, and you can get really risk averse,

172
00:09:38,320 --> 00:09:41,520
Speaker 4: meaning you feel like Hey, this is a successful business.

173
00:09:41,559 --> 00:09:43,280
Speaker 4: I don't want to do anything to mess it up,

174
00:09:43,600 --> 00:09:45,200
Speaker 4: so I don't need to try new things. Well, that's

175
00:09:45,280 --> 00:09:48,520
Speaker 4: exactly the recipe to kind of be shrinking, and infrastructure

176
00:09:48,559 --> 00:09:50,840
Speaker 4: had been shrinking for a little while, and so a

177
00:09:50,880 --> 00:09:54,480
Speaker 4: lot of what the challenge was for me was to

178
00:09:54,520 --> 00:09:58,720
Speaker 4: invigorate that risk taking and get to a growth mindset

179
00:09:58,720 --> 00:10:01,360
Speaker 4: where you're trying new things and seeing what works and

180
00:10:01,400 --> 00:10:03,720
Speaker 4: what doesn't work, and changing some of the models, like

181
00:10:04,040 --> 00:10:06,720
Speaker 4: investing a little bit less in hardware for some software

182
00:10:06,720 --> 00:10:10,480
Speaker 4: differentiation that goes into the hardware. So it's been very

183
00:10:10,480 --> 00:10:12,960
Speaker 4: successful so far, and it's been a good journey. It's

184
00:10:12,960 --> 00:10:14,160
Speaker 4: almost four years now.

185
00:10:14,559 --> 00:10:17,480
Speaker 3: Give me an example of what was a really hard

186
00:10:17,520 --> 00:10:19,240
Speaker 3: problem that you've dealt with in those four years.

187
00:10:19,240 --> 00:10:22,160
Speaker 4: So, boy, a really hard problem.

188
00:10:22,240 --> 00:10:25,200
Speaker 3: An interesting and are you interesting is a better word

189
00:10:25,240 --> 00:10:25,560
Speaker 3: than art.

190
00:10:25,840 --> 00:10:27,840
Speaker 4: One of the first things that I kind of chewed

191
00:10:27,880 --> 00:10:30,520
Speaker 4: on a little bit is I talked about how we

192
00:10:30,559 --> 00:10:33,679
Speaker 4: have Z power and storage. The Z and power product

193
00:10:33,720 --> 00:10:36,480
Speaker 4: lines are well known in the industry. Is is really

194
00:10:36,520 --> 00:10:39,840
Speaker 4: fit for purpose computing that have strengths that you know

195
00:10:39,960 --> 00:10:43,880
Speaker 4: Z runs you know most of the world's economic backbone,

196
00:10:43,920 --> 00:10:47,520
Speaker 4: and if you use a credit card. Ninety percent of

197
00:10:47,559 --> 00:10:51,040
Speaker 4: credit card transactions for the globe go through these Z

198
00:10:51,520 --> 00:10:54,080
Speaker 4: mainframes there in every bank there. You know, it's a

199
00:10:54,080 --> 00:10:56,880
Speaker 4: big business. It's well known in the industry. Same with power,

200
00:10:57,320 --> 00:11:02,240
Speaker 4: very tuned and optimized for smaller operations than our giant

201
00:11:02,360 --> 00:11:08,319
Speaker 4: Z mainframes, but really mission critical workloads for retail, for insurance,

202
00:11:08,320 --> 00:11:11,520
Speaker 4: for banking, for all of that. Our storage business not

203
00:11:11,640 --> 00:11:14,439
Speaker 4: so well known. In fact, when I came I thought,

204
00:11:14,679 --> 00:11:17,079
Speaker 4: did they have storage? Well, I have storage when I

205
00:11:17,240 --> 00:11:19,680
Speaker 4: come in too. I So I got online and I thought,

206
00:11:19,760 --> 00:11:21,560
Speaker 4: it's still hard for me to tell did they have

207
00:11:21,679 --> 00:11:24,000
Speaker 4: storage or not? Now I own a storage business. So

208
00:11:24,080 --> 00:11:25,640
Speaker 4: one of the things was not just to get the

209
00:11:25,679 --> 00:11:29,280
Speaker 4: market perception up, but to invest in that business. Because

210
00:11:29,320 --> 00:11:32,000
Speaker 4: if you look at infrastructure overall around the globe, it's

211
00:11:32,040 --> 00:11:35,600
Speaker 4: growing at five percent a year. The infrastructure business had

212
00:11:35,640 --> 00:11:38,320
Speaker 4: been kind of flat to declining, and so a challenge

213
00:11:38,360 --> 00:11:40,319
Speaker 4: was how do we grab onto the growth. Well, one

214
00:11:40,320 --> 00:11:43,040
Speaker 4: of the biggest growth areas due to the explosion of

215
00:11:43,120 --> 00:11:45,720
Speaker 4: data in the world is storage, So what do you

216
00:11:45,760 --> 00:11:47,960
Speaker 4: do to kind of get on that growth rate. So

217
00:11:47,960 --> 00:11:51,600
Speaker 4: we did a lot of reinvigoration of the innovation in

218
00:11:51,640 --> 00:11:54,520
Speaker 4: that a lot of software value, add a lot of

219
00:11:54,840 --> 00:11:57,920
Speaker 4: doubling down on the things that are working. Portfolio rationalization

220
00:11:58,000 --> 00:12:00,200
Speaker 4: where you segment the market and you say, okay, we're

221
00:12:00,360 --> 00:12:02,160
Speaker 4: going to do less of this and really go big

222
00:12:02,200 --> 00:12:05,079
Speaker 4: in these areas. And that's been probably the most dramatic

223
00:12:05,080 --> 00:12:08,440
Speaker 4: turnaround inside the group. Is our storage thing. When you

224
00:12:08,480 --> 00:12:11,200
Speaker 4: say it's a hard problem, it's not just oh, you know,

225
00:12:11,240 --> 00:12:15,800
Speaker 4: how do we do the math? No, it's cultural. It's strategy,

226
00:12:15,880 --> 00:12:18,040
Speaker 4: and how do you get the strategy set. It's segmentation,

227
00:12:18,160 --> 00:12:21,280
Speaker 4: it's product strategy at a granular level across a bunch

228
00:12:21,320 --> 00:12:24,640
Speaker 4: of dimensions, and then putting the investment behind it. It's

229
00:12:24,679 --> 00:12:26,880
Speaker 4: a big challenge. It takes a long time, but it's working,

230
00:12:26,960 --> 00:12:28,920
Speaker 4: so we're happy with Yeah.

231
00:12:28,600 --> 00:12:30,559
Speaker 3: Tell me give me a little bit of perspective on

232
00:12:31,120 --> 00:12:34,920
Speaker 3: you've been there four years. Imagine we're having this conversation

233
00:12:35,000 --> 00:12:35,640
Speaker 3: four years ago.

234
00:12:36,120 --> 00:12:38,280
Speaker 4: Yeah, what sorts.

235
00:12:38,080 --> 00:12:40,200
Speaker 3: Of things have happened over the last four years that

236
00:12:41,200 --> 00:12:43,680
Speaker 3: have surprised you that you didn't see come? At least

237
00:12:44,480 --> 00:12:46,600
Speaker 3: we had exactly the same conversation four years ago.

238
00:12:48,000 --> 00:12:49,720
Speaker 4: No, because I didn't know what was in I'll tell

239
00:12:49,720 --> 00:12:53,559
Speaker 4: you some of the biggest surprises I thought from the outside,

240
00:12:53,960 --> 00:12:57,199
Speaker 4: and you know, you hear from a lot of customers,

241
00:12:57,360 --> 00:13:00,560
Speaker 4: especially ten years ago, we're all going to we're all

242
00:13:00,840 --> 00:13:03,360
Speaker 4: so I thought, well, I wonder if the mainframe business

243
00:13:03,400 --> 00:13:06,160
Speaker 4: is struggling. When I get inside of there, I found

244
00:13:06,200 --> 00:13:08,720
Speaker 4: the opposite to be true. The mainframe business is actually

245
00:13:08,760 --> 00:13:12,839
Speaker 4: flourishing because transaction demand across the globe has done nothing

246
00:13:12,920 --> 00:13:16,120
Speaker 4: but grow. And even more surprising was the level of

247
00:13:16,200 --> 00:13:19,840
Speaker 4: innovation that the team was already doing in mainframes before

248
00:13:19,880 --> 00:13:24,400
Speaker 4: I got here was astounding. For example, we have AI.

249
00:13:24,880 --> 00:13:29,360
Speaker 4: They were building AI technology into the mainframe processors three

250
00:13:29,440 --> 00:13:32,720
Speaker 4: years before chat GPT made everybody talk about it in

251
00:13:32,760 --> 00:13:36,840
Speaker 4: the industry, So that was really pleasantly surprising. So that

252
00:13:37,000 --> 00:13:41,840
Speaker 4: was wonderful. Other surprises I knew about the kind of

253
00:13:41,840 --> 00:13:44,960
Speaker 4: the IP of IBM and the mystique in that, and

254
00:13:45,040 --> 00:13:47,520
Speaker 4: I used to joke with people, especially on the outside,

255
00:13:47,520 --> 00:13:48,960
Speaker 4: I said, I can't wait to get in there and

256
00:13:48,960 --> 00:13:51,800
Speaker 4: see what's in the big blue toolbox? Right, what are

257
00:13:51,840 --> 00:13:55,160
Speaker 4: all the things they have going on? I way underestimated

258
00:13:55,360 --> 00:13:57,720
Speaker 4: the size of the big blue toolbox and what was

259
00:13:57,760 --> 00:14:02,640
Speaker 4: in their meaning amount of really hardcore research that we're

260
00:14:02,679 --> 00:14:06,040
Speaker 4: still doing into how to build chips and how to

261
00:14:06,080 --> 00:14:09,079
Speaker 4: get to things beyond two nanimeter and that kind of

262
00:14:09,160 --> 00:14:15,160
Speaker 4: capability packaging industry leading packaging technologies, and that's in my

263
00:14:15,320 --> 00:14:18,079
Speaker 4: hardware kind of patch quantum. The next thing that will

264
00:14:18,120 --> 00:14:22,720
Speaker 4: come after we're done talking about AI. You know, all

265
00:14:22,760 --> 00:14:25,560
Speaker 4: of those things were surprising, But it wasn't just that.

266
00:14:25,640 --> 00:14:28,200
Speaker 4: It was then the software innovations that are going on

267
00:14:28,360 --> 00:14:33,160
Speaker 4: heavy investment in AI technologies before it was really popular

268
00:14:33,200 --> 00:14:35,840
Speaker 4: to be talking about that. But as I saw that,

269
00:14:35,920 --> 00:14:38,560
Speaker 4: I thought this is going to get really fun. Because

270
00:14:38,600 --> 00:14:41,080
Speaker 4: I had a good feel for where the industry was going.

271
00:14:41,920 --> 00:14:44,440
Speaker 4: I just didn't and I knew, man, I know that

272
00:14:44,560 --> 00:14:47,200
Speaker 4: talent is really good and there's brilliant people there, but

273
00:14:47,240 --> 00:14:51,160
Speaker 4: I didn't know the level of IP frankly that IBM

274
00:14:51,280 --> 00:14:54,200
Speaker 4: had at its disposal. And now you're seeing that in

275
00:14:54,240 --> 00:14:57,480
Speaker 4: things like Watson X and things like AI in mainframes,

276
00:14:57,480 --> 00:14:57,880
Speaker 4: et cetera.

277
00:14:58,280 --> 00:15:01,520
Speaker 3: Building on that. Since you put AI, can you walk

278
00:15:01,560 --> 00:15:05,360
Speaker 3: me through what has to happen from your perspective, from

279
00:15:05,400 --> 00:15:11,280
Speaker 3: the infrastructure perspective to make the AI explosion work? Yeah,

280
00:15:11,320 --> 00:15:13,320
Speaker 3: so everyone wants to do more of this stuff. Yes,

281
00:15:13,520 --> 00:15:16,000
Speaker 3: clearly there has to be some underpinning of it.

282
00:15:16,440 --> 00:15:20,000
Speaker 4: Yeah, I would tell you, you know, I think that

283
00:15:20,160 --> 00:15:22,440
Speaker 4: people feel like where we're at right now in the

284
00:15:22,480 --> 00:15:25,560
Speaker 4: AI journey had to do with one specific piece of software.

285
00:15:25,920 --> 00:15:30,040
Speaker 4: I think the inflection point for that whole thing really

286
00:15:30,960 --> 00:15:34,680
Speaker 4: at its root was around hardware, meaning the algorithms needed

287
00:15:34,720 --> 00:15:37,280
Speaker 4: to do larger language models. And all of that had

288
00:15:37,360 --> 00:15:40,000
Speaker 4: been around, they'd been talked about in the industry, but

289
00:15:40,040 --> 00:15:43,080
Speaker 4: at some point you hit a tipping point of hardware

290
00:15:43,120 --> 00:15:45,760
Speaker 4: capability where it's like, oh, now we can do this

291
00:15:45,880 --> 00:15:49,280
Speaker 4: in a broof force way, massive amounts of matrix math

292
00:15:49,360 --> 00:15:52,320
Speaker 4: to get weights correct so that you can do you know,

293
00:15:52,360 --> 00:15:55,680
Speaker 4: the right level of predictions that enable large language models.

294
00:15:56,040 --> 00:15:58,600
Speaker 4: And once we got to that horsepower, and that's why

295
00:15:58,640 --> 00:16:01,240
Speaker 4: you hear about giant g pus that are driving this

296
00:16:01,600 --> 00:16:03,720
Speaker 4: and the sales of those, et cetera. It's because we

297
00:16:03,920 --> 00:16:05,760
Speaker 4: just barely got over the hump where you can do

298
00:16:05,800 --> 00:16:10,440
Speaker 4: these big hard things in terms of hardware capability to

299
00:16:10,600 --> 00:16:11,320
Speaker 4: do it.

300
00:16:11,440 --> 00:16:14,280
Speaker 3: Give me a layman, give me a sense of when

301
00:16:14,320 --> 00:16:16,480
Speaker 3: you say there was a kind of threshold where suddenly

302
00:16:16,480 --> 00:16:17,640
Speaker 3: these things became possible.

303
00:16:17,760 --> 00:16:20,560
Speaker 4: Yeah, I don't know if there's an exact number, but

304
00:16:21,720 --> 00:16:23,800
Speaker 4: and more basic question that I get from a lot

305
00:16:23,840 --> 00:16:26,520
Speaker 4: of people, you know, my friends and family outside. Is

306
00:16:26,840 --> 00:16:31,560
Speaker 4: why GPUs. What does a GPU, a graphics processor have

307
00:16:31,680 --> 00:16:36,080
Speaker 4: to do with AI. It's not, Well, graphics processors are

308
00:16:36,120 --> 00:16:39,880
Speaker 4: really good at this thing matrix math, because they're figuring

309
00:16:39,920 --> 00:16:43,280
Speaker 4: out how do I map a pixel? And as I

310
00:16:43,400 --> 00:16:47,560
Speaker 4: move an object across the screen, it's essentially matrix math

311
00:16:47,640 --> 00:16:50,320
Speaker 4: to figure out, Okay, what does what does this pixel

312
00:16:50,400 --> 00:16:53,120
Speaker 4: on a screen look like? And what it's doing? And

313
00:16:53,160 --> 00:16:56,480
Speaker 4: as you you know, we've gotten more high resolution graphics,

314
00:16:56,520 --> 00:16:59,320
Speaker 4: more high resolution monitors, et cetera. It's a lot more

315
00:16:59,320 --> 00:17:01,240
Speaker 4: pixels and a lot more math and a lot more

316
00:17:01,280 --> 00:17:04,359
Speaker 4: matrix math about how you compute that. The first big

317
00:17:04,359 --> 00:17:06,679
Speaker 4: thing that kind of started to look like that, it

318
00:17:06,720 --> 00:17:10,199
Speaker 4: turns out, was crypto and crypto mining, and so you

319
00:17:10,240 --> 00:17:13,959
Speaker 4: saw graphics companies starting to sell to crypto. The technology

320
00:17:13,960 --> 00:17:15,960
Speaker 4: got to a certain point and there were use cases

321
00:17:16,000 --> 00:17:18,480
Speaker 4: like bitcoin in that that kind of said, hey, we

322
00:17:18,520 --> 00:17:20,720
Speaker 4: need to do a lot of this matrix math to

323
00:17:20,760 --> 00:17:23,440
Speaker 4: be able to do that. So graphic chips were a

324
00:17:23,520 --> 00:17:26,440
Speaker 4: natural fit and that kind of sustain But meanwhile, behind

325
00:17:26,480 --> 00:17:29,320
Speaker 4: the scenes, a lot of this AI AI is about

326
00:17:30,000 --> 00:17:34,439
Speaker 4: numeric calculations having to do with weights and matrices that say,

327
00:17:34,880 --> 00:17:38,480
Speaker 4: you know, giant consolidated things that predict what's going to

328
00:17:38,680 --> 00:17:41,159
Speaker 4: kind of happen based on what other things have happened,

329
00:17:41,200 --> 00:17:44,880
Speaker 4: just like predicting where pixel goes. But it's really about

330
00:17:45,480 --> 00:17:48,560
Speaker 4: being able to do enough data in jest to be

331
00:17:48,560 --> 00:17:51,160
Speaker 4: able to do and then the calculations to be able

332
00:17:51,200 --> 00:17:55,640
Speaker 4: to simplify things like entire sets of language or giant

333
00:17:55,920 --> 00:17:58,640
Speaker 4: chunks of the Internet, to get enough weightings in there

334
00:17:58,640 --> 00:18:01,280
Speaker 4: to be able to say, okay, we can predict what

335
00:18:01,320 --> 00:18:04,720
Speaker 4: you would say in this language based on all of

336
00:18:04,720 --> 00:18:07,359
Speaker 4: the volumes of stuff that we've seen that when you

337
00:18:07,520 --> 00:18:10,120
Speaker 4: start talking like this, the next word is likely, oh

338
00:18:10,160 --> 00:18:11,040
Speaker 4: it's this. Yeah.

339
00:18:11,080 --> 00:18:13,320
Speaker 3: So, But my point is to get to that point,

340
00:18:13,440 --> 00:18:17,240
Speaker 3: that's threshold. We got there because was it because we

341
00:18:17,320 --> 00:18:20,159
Speaker 3: simply threw a lot more resources at the problem or

342
00:18:20,240 --> 00:18:24,480
Speaker 3: is it because the underlying technology got suddenly or gradually

343
00:18:24,640 --> 00:18:25,800
Speaker 3: so much more efficient.

344
00:18:25,920 --> 00:18:29,160
Speaker 4: It's always yes and yes. But you know, the industry

345
00:18:29,240 --> 00:18:31,400
Speaker 4: for a lot of years would talk about Moore's law.

346
00:18:31,880 --> 00:18:35,639
Speaker 3: Well, quick, will you define for us More's law for

347
00:18:35,680 --> 00:18:37,240
Speaker 3: those of those who's forgotten it.

348
00:18:37,480 --> 00:18:40,640
Speaker 4: Yeah, So Gordon Moore at Intel coined this thing. It

349
00:18:40,680 --> 00:18:45,399
Speaker 4: was basically that the horsepower I'm going to translate it

350
00:18:45,520 --> 00:18:51,120
Speaker 4: roughly of technology will double every couple of years. We're

351
00:18:51,160 --> 00:18:53,840
Speaker 4: still on Moore's law. Moore's law changed a little bit.

352
00:18:54,240 --> 00:18:57,080
Speaker 4: For a while, it was always about frequency. Things would

353
00:18:57,080 --> 00:19:00,280
Speaker 4: go faster, faster, faster. That kind of petered out. But

354
00:19:00,359 --> 00:19:03,560
Speaker 4: what happened is, rather than faster, faster, faster, we did

355
00:19:03,560 --> 00:19:06,840
Speaker 4: more and more and more. So rather than one operating

356
00:19:06,960 --> 00:19:10,800
Speaker 4: unit going a lot faster on its throughput, you put

357
00:19:10,840 --> 00:19:13,280
Speaker 4: ten operating units on a chip, now you put one

358
00:19:13,359 --> 00:19:16,399
Speaker 4: hundred operating units on a chip, now a thousand. Some

359
00:19:16,480 --> 00:19:21,120
Speaker 4: of these problems, the matrix math problems scale parallel extremely well.

360
00:19:21,119 --> 00:19:23,399
Speaker 4: You don't have to do something really fast, you just

361
00:19:23,520 --> 00:19:25,520
Speaker 4: have to do a lot of the similar things in

362
00:19:25,560 --> 00:19:28,520
Speaker 4: parallel at the same time. So again that kind of

363
00:19:28,520 --> 00:19:31,240
Speaker 4: that extension of Moore's law, more and more hardware on

364
00:19:31,320 --> 00:19:32,720
Speaker 4: a chip to be able to do more and more

365
00:19:32,720 --> 00:19:35,960
Speaker 4: of those calculations in parallel and come up with it.

366
00:19:36,080 --> 00:19:39,200
Speaker 3: And we said, yeah, was that threshold predictable? In other words,

367
00:19:39,200 --> 00:19:42,080
Speaker 3: see people in the industry, like you sit down X

368
00:19:42,119 --> 00:19:44,160
Speaker 3: number of years ago and say, when we get here,

369
00:19:45,160 --> 00:19:48,200
Speaker 3: AI is going to become much more of a It's funny.

370
00:19:48,760 --> 00:19:55,159
Speaker 4: The horsepower that very predictable, the use cases not always

371
00:19:55,240 --> 00:19:57,639
Speaker 4: so easy to kind of figure out. That's where the

372
00:19:58,080 --> 00:20:01,199
Speaker 4: human spirit kind of gets involved. I think for some

373
00:20:01,240 --> 00:20:03,600
Speaker 4: people that say, oh, I saw that coming, but people

374
00:20:03,640 --> 00:20:07,280
Speaker 4: have been predicting kind of the rise of AI for

375
00:20:07,640 --> 00:20:09,600
Speaker 4: twenty five years. Oh well, then when we get to

376
00:20:09,600 --> 00:20:11,800
Speaker 4: this next gener oh when we get here, it kind

377
00:20:11,800 --> 00:20:15,600
Speaker 4: of hadn't happened. There's always a magic point where you

378
00:20:15,720 --> 00:20:18,040
Speaker 4: kind of get to where the technology and the use

379
00:20:18,080 --> 00:20:20,479
Speaker 4: case and somebody does something to kind of make it

380
00:20:21,280 --> 00:20:23,120
Speaker 4: catch on. And I think we're at one of those

381
00:20:23,119 --> 00:20:25,120
Speaker 4: moments in AI for sure right now. And I don't

382
00:20:25,160 --> 00:20:27,080
Speaker 4: think it's you know, people that have said, oh, this

383
00:20:27,200 --> 00:20:30,399
Speaker 4: is just the latest wave of you know, I hear

384
00:20:30,680 --> 00:20:33,199
Speaker 4: I've heard this about a lot of technologies, but AI

385
00:20:33,320 --> 00:20:35,720
Speaker 4: is the technology the future, and it always will be.

386
00:20:35,960 --> 00:20:38,840
Speaker 4: I used to hear that. You're not hearing that now, right,

387
00:20:38,880 --> 00:20:42,560
Speaker 4: It's like, no, it's primetime. It will change everything, just

388
00:20:42,720 --> 00:20:44,760
Speaker 4: like some of these other things changed everything.

389
00:20:45,000 --> 00:20:49,439
Speaker 3: I noticed it if personally when I speak somewhere or

390
00:20:49,480 --> 00:20:53,359
Speaker 3: I'm listening an audience somewhere. Over the last let's say

391
00:20:53,480 --> 00:20:58,600
Speaker 3: twelve months, there's always a whole bunch of AI questions. Yes,

392
00:20:58,760 --> 00:21:00,960
Speaker 3: And if I go back to years ago, there were

393
00:21:00,960 --> 00:21:01,840
Speaker 3: no AI questions.

394
00:21:01,920 --> 00:21:02,120
Speaker 4: Yes.

395
00:21:02,720 --> 00:21:05,040
Speaker 3: Now my question is, so there's been this explosion on

396
00:21:05,080 --> 00:21:09,480
Speaker 3: the in popular fascination with what's going on AI. It

397
00:21:09,560 --> 00:21:12,560
Speaker 3: seems like the last year. I agree with you in

398
00:21:12,640 --> 00:21:18,920
Speaker 3: your world, when did the explosion of conversation around this start?

399
00:21:19,640 --> 00:21:28,960
Speaker 4: It's I love this question because IBM had a fairly

400
00:21:29,119 --> 00:21:34,199
Speaker 4: big effort and business called Watson before Watson X. And

401
00:21:34,240 --> 00:21:37,160
Speaker 4: this is going back kind of ten years. I'll give

402
00:21:37,160 --> 00:21:40,000
Speaker 4: you another kind of example. I knew about a lot

403
00:21:40,040 --> 00:21:43,479
Speaker 4: of tablet technology before there was an iPad, a lot.

404
00:21:43,680 --> 00:21:46,040
Speaker 4: For ten years, there were a lot, but it kind

405
00:21:46,040 --> 00:21:49,720
Speaker 4: of takes a magic combination of the technology, the user experienced,

406
00:21:49,760 --> 00:21:52,439
Speaker 4: the software, and the need and the market ready for

407
00:21:52,480 --> 00:21:54,680
Speaker 4: it to kind of go. Now it's the thing. Now

408
00:21:54,680 --> 00:21:56,800
Speaker 4: we all have either an iPad or we have the

409
00:21:57,119 --> 00:22:00,119
Speaker 4: Google equivalent Tom and so I think this is a

410
00:22:00,160 --> 00:22:03,080
Speaker 4: little like that, meaning IBM was on the right track

411
00:22:03,119 --> 00:22:06,040
Speaker 4: with Watson. Some of the hardware wasn't there, the use

412
00:22:06,080 --> 00:22:08,720
Speaker 4: cases weren't exactly figured out. Some of the early use

413
00:22:08,720 --> 00:22:11,840
Speaker 4: cases didn't pan out. Perfectly. But the good news about

414
00:22:11,880 --> 00:22:15,359
Speaker 4: that is it's back to that culture of risk taking.

415
00:22:15,440 --> 00:22:17,920
Speaker 4: You don't look back on that and say, oh, we

416
00:22:17,920 --> 00:22:19,480
Speaker 4: shouldn't have done that, that was a bad idea. I

417
00:22:19,480 --> 00:22:21,040
Speaker 4: know you look back on that and say, what did

418
00:22:21,080 --> 00:22:23,480
Speaker 4: we learn? How should we try something new? How would

419
00:22:23,480 --> 00:22:25,879
Speaker 4: we pivot this time? That's what we've done with Watson

420
00:22:26,080 --> 00:22:29,719
Speaker 4: X and now that's a growing, healthy piece of our

421
00:22:29,760 --> 00:22:32,719
Speaker 4: business and very important our strategic sure, so we're all in.

422
00:22:33,359 --> 00:22:39,879
Speaker 3: I've always investigated by the gap between insider sense of

423
00:22:39,920 --> 00:22:42,120
Speaker 3: what is happening in an outsider sense, like.

424
00:22:42,200 --> 00:22:45,160
Speaker 4: It absolutely is that in this case, we've all been

425
00:22:45,240 --> 00:22:49,240
Speaker 4: talking about and thinking about AI and is it time

426
00:22:49,280 --> 00:22:52,040
Speaker 4: for that and what does this mean? Et cetera. And

427
00:22:52,119 --> 00:22:54,880
Speaker 4: yet none of us really predicted that actual moment, which

428
00:22:54,920 --> 00:22:58,560
Speaker 4: is kind of you know, early twenty twenty two where

429
00:22:58,600 --> 00:23:02,600
Speaker 4: it was like, oh, now you have a simple human

430
00:23:02,640 --> 00:23:08,400
Speaker 4: interface of software innovation combined with large language models. There's

431
00:23:08,440 --> 00:23:11,639
Speaker 4: a moment there where you're like, oh, Unlike, you know,

432
00:23:11,640 --> 00:23:13,320
Speaker 4: I think all of us are frustrated if we ask

433
00:23:13,359 --> 00:23:15,760
Speaker 4: our phone, hey, tell me about this, and it says

434
00:23:16,320 --> 00:23:18,280
Speaker 4: I found this on the web page. That does you

435
00:23:18,400 --> 00:23:21,680
Speaker 4: no good? But you know, all of a sudden, with chat,

436
00:23:21,760 --> 00:23:24,080
Speaker 4: GPT and some of these other things, you could ask

437
00:23:24,119 --> 00:23:26,400
Speaker 4: a question, it would give you a clear answer. Sometimes

438
00:23:26,520 --> 00:23:28,960
Speaker 4: is wrong, but at least it was like I'm getting

439
00:23:28,960 --> 00:23:30,720
Speaker 4: an answer rather than hey, I don't know if there's

440
00:23:30,760 --> 00:23:34,200
Speaker 4: some references. Good luck to you. And that's really changing.

441
00:23:34,760 --> 00:23:39,240
Speaker 3: Talk about the kind of macro trends that are going

442
00:23:39,320 --> 00:23:42,359
Speaker 3: to shape your infrastructure battle.

443
00:23:42,480 --> 00:23:45,359
Speaker 4: Yeah, we've talked about if you already, but I'm actually

444
00:23:45,400 --> 00:23:49,320
Speaker 4: going to go a little different direction. So macro trans first,

445
00:23:49,800 --> 00:23:54,040
Speaker 4: and this one has been before even even this AI conversation,

446
00:23:54,119 --> 00:23:59,280
Speaker 4: that we've had explosion of data. As humans, we don't

447
00:23:59,400 --> 00:24:04,639
Speaker 4: think exponentially very well. We really struggle with exponential thinking.

448
00:24:04,840 --> 00:24:07,199
Speaker 4: We think linearly, Oh, there'll be more, there'll be more,

449
00:24:07,240 --> 00:24:09,440
Speaker 4: they'll be more, but we don't think well when it's

450
00:24:09,440 --> 00:24:11,440
Speaker 4: like no, they'll be more, and they'll be ten times more,

451
00:24:11,440 --> 00:24:13,560
Speaker 4: and then there'll be ten times that more. That's what's

452
00:24:13,600 --> 00:24:16,640
Speaker 4: going on with data right now in our industry. It's

453
00:24:16,680 --> 00:24:19,080
Speaker 4: one of the reasons that that storage business is doing

454
00:24:19,080 --> 00:24:21,240
Speaker 4: so well is they're just more and more and more data.

455
00:24:22,840 --> 00:24:24,680
Speaker 4: You know, you'd say, well, how can there be more data?

456
00:24:24,720 --> 00:24:27,119
Speaker 4: It's just life and that thing. The things that we

457
00:24:27,200 --> 00:24:30,760
Speaker 4: care about, video capture, video images, you know, the the

458
00:24:31,640 --> 00:24:34,280
Speaker 4: you I don't know from my parents, you needed a

459
00:24:34,440 --> 00:24:37,440
Speaker 4: drawer with all your family photos. Now we need gigabytes

460
00:24:37,440 --> 00:24:38,000
Speaker 4: and gigabytes.

461
00:24:38,080 --> 00:24:40,399
Speaker 3: You knew how many pictures my wife has taken off

462
00:24:40,400 --> 00:24:44,720
Speaker 3: our children, you would exactly exactly, So that's your case.

463
00:24:44,800 --> 00:24:47,480
Speaker 4: Now. Think of companies who used to just think about

464
00:24:47,480 --> 00:24:50,919
Speaker 4: their transaction data. What's the ledger say that now have

465
00:24:51,160 --> 00:24:54,679
Speaker 4: video assets of all of their campaigns and their marketing.

466
00:24:54,720 --> 00:24:57,359
Speaker 4: They're trying to figure out, you know, what campaigns are

467
00:24:57,400 --> 00:24:59,920
Speaker 4: working the best. So it's just an explosion of data

468
00:25:00,240 --> 00:25:03,439
Speaker 4: and that's not going to stop. Dealing with that, and

469
00:25:03,480 --> 00:25:08,840
Speaker 4: more importantly, getting value from that data is a massive

470
00:25:08,960 --> 00:25:13,280
Speaker 4: trend in the industry. Second trend AI, and this is

471
00:25:13,480 --> 00:25:15,920
Speaker 4: the AI. Not like we were just talking about about

472
00:25:15,960 --> 00:25:18,119
Speaker 4: how it changes how I search for things or how

473
00:25:18,119 --> 00:25:22,160
Speaker 4: I learn about things. But I would argue, dealing with

474
00:25:22,200 --> 00:25:24,560
Speaker 4: that data, how do I figure out what's in all

475
00:25:24,600 --> 00:25:27,520
Speaker 4: those video streams? How do I figure out Okay, I

476
00:25:27,560 --> 00:25:30,600
Speaker 4: want all of the chunks of my corporate video that

477
00:25:30,720 --> 00:25:35,120
Speaker 4: have to do with client buying some specific product or something.

478
00:25:35,160 --> 00:25:38,119
Speaker 4: That's a different problem. It's not just okay, we'll look

479
00:25:38,160 --> 00:25:40,639
Speaker 4: it up in a spreadsheet and here's the math associated

480
00:25:40,680 --> 00:25:43,880
Speaker 4: with that. That is a huge trend in the industry.

481
00:25:43,920 --> 00:25:46,000
Speaker 4: You're seeing it play out in this regard, it's a

482
00:25:46,000 --> 00:25:50,320
Speaker 4: little different bent on AI. Fraud detection is the one

483
00:25:50,320 --> 00:25:53,200
Speaker 4: that we cite in our mainframes. It's a similar problem

484
00:25:53,240 --> 00:25:56,240
Speaker 4: where it was kind of a traditional AI problem. Look

485
00:25:56,320 --> 00:25:59,760
Speaker 4: up a rule. You know, if somebody does two small

486
00:25:59,760 --> 00:26:02,399
Speaker 4: trend actions than a massive one, it might be fraud,

487
00:26:02,440 --> 00:26:05,480
Speaker 4: right because they were seeing whether it were now to

488
00:26:05,560 --> 00:26:09,679
Speaker 4: detect fraud, you might be saying, okay to transactions then

489
00:26:09,720 --> 00:26:14,280
Speaker 4: a huge one. Plus does this entity have a real address? Second,

490
00:26:14,400 --> 00:26:17,520
Speaker 4: is there any web traffic on you know, better Business

491
00:26:17,520 --> 00:26:19,280
Speaker 4: Bureau kind of things that says this is a bad

492
00:26:19,320 --> 00:26:21,639
Speaker 4: business that can help you with fraud. So it's a

493
00:26:21,680 --> 00:26:24,679
Speaker 4: lot more of a it's an EXPERTI problem. It's a

494
00:26:24,720 --> 00:26:27,359
Speaker 4: holistic problem that it takes a lot more than just

495
00:26:27,840 --> 00:26:30,760
Speaker 4: you know, little chunks of rules, et cetera. And then

496
00:26:30,800 --> 00:26:34,280
Speaker 4: the third one you know after AI, is the nature

497
00:26:34,320 --> 00:26:38,120
Speaker 4: of hybrid it or hybrid computing. For a while ten

498
00:26:38,200 --> 00:26:40,960
Speaker 4: years ago, when cloud was on the rise, I think

499
00:26:41,359 --> 00:26:44,679
Speaker 4: the notion of hybrid computing basically having to do with

500
00:26:44,800 --> 00:26:48,000
Speaker 4: things in the cloud versus things that people still have

501
00:26:49,040 --> 00:26:52,159
Speaker 4: on the premises inside of business. It was almost a

502
00:26:52,200 --> 00:26:56,080
Speaker 4: religious argument. Now it's no, it's the reality. And the

503
00:26:56,160 --> 00:26:59,520
Speaker 4: reason is because that data that I talked about is

504
00:26:59,560 --> 00:27:03,560
Speaker 4: the life blood of these companies, particularly IBM's companies are

505
00:27:03,760 --> 00:27:07,400
Speaker 4: clients that usually that data has to be secure, they

506
00:27:07,440 --> 00:27:09,960
Speaker 4: have to be able to get value from it. It

507
00:27:10,040 --> 00:27:11,840
Speaker 4: is the lifeblood of the company. If you go to

508
00:27:11,880 --> 00:27:14,679
Speaker 4: an ATM and you can't get your money out to

509
00:27:14,720 --> 00:27:18,680
Speaker 4: our financial transactions, if that lasts a day, you're probably

510
00:27:18,720 --> 00:27:21,119
Speaker 4: going to change banks immediately. So it's like life or

511
00:27:21,160 --> 00:27:27,800
Speaker 4: death for these companies. So having that hybrid infrastructure so

512
00:27:27,840 --> 00:27:30,760
Speaker 4: that they can still hold their data, you still interact

513
00:27:30,800 --> 00:27:33,520
Speaker 4: with clouds and still get value from it from AI.

514
00:27:34,040 --> 00:27:38,080
Speaker 4: That's kind of the magic where we play, and it's

515
00:27:38,119 --> 00:27:41,760
Speaker 4: a huge business opportunity. It is a true inflection point

516
00:27:41,800 --> 00:27:44,679
Speaker 4: for the industry. I'm going to go back.

517
00:27:45,760 --> 00:27:47,639
Speaker 3: I interrupted you when you were in the middle of

518
00:27:47,680 --> 00:27:50,560
Speaker 3: a rellion. We were talking about what has to happen

519
00:27:50,680 --> 00:27:56,040
Speaker 3: for AI to scale from the infrastructure standpoint. You gave

520
00:27:56,080 --> 00:27:58,399
Speaker 3: one example that I got you off on a tangent.

521
00:27:58,760 --> 00:28:02,360
Speaker 3: Can you go back and talk very so practically, like, so,

522
00:28:02,480 --> 00:28:04,760
Speaker 3: I'm you know, I'm a big company. I have all

523
00:28:04,760 --> 00:28:08,000
Speaker 3: these dreams of AI, of how I'm going to use

524
00:28:08,040 --> 00:28:11,240
Speaker 3: this dratically, So give me a very granular sense of

525
00:28:11,640 --> 00:28:14,360
Speaker 3: the works you have to do, yeah, to make that

526
00:28:14,480 --> 00:28:15,240
Speaker 3: dream possible.

527
00:28:15,680 --> 00:28:18,919
Speaker 4: So let me first say what the company has to do,

528
00:28:18,960 --> 00:28:20,920
Speaker 4: and then maybe I'll say, then how do I help them?

529
00:28:21,000 --> 00:28:23,159
Speaker 4: If that makes sense? So if I'm a company and

530
00:28:23,160 --> 00:28:25,159
Speaker 4: I want to do that, So it turns out I

531
00:28:25,200 --> 00:28:29,000
Speaker 4: am a company meaning I want to use AI in

532
00:28:29,040 --> 00:28:32,879
Speaker 4: my processes. I mentioned that I have a global network

533
00:28:32,920 --> 00:28:37,080
Speaker 4: of thirteen thousand employees that support our infrastructure around the world.

534
00:28:37,480 --> 00:28:42,840
Speaker 4: That challenge is a great challenge for AI. That means

535
00:28:42,920 --> 00:28:47,480
Speaker 4: I have data for every customer situation for thirteen thousand

536
00:28:47,520 --> 00:28:50,960
Speaker 4: employees globally around the world on what was their problem,

537
00:28:51,000 --> 00:28:54,480
Speaker 4: how did we fix it, what next steps did they

538
00:28:54,480 --> 00:28:56,920
Speaker 4: have to do, how did they remediate that? That data

539
00:28:57,000 --> 00:28:59,480
Speaker 4: is extremely valuable to me because if I can get

540
00:28:59,480 --> 00:29:02,080
Speaker 4: better at doing that than anybody else in the world,

541
00:29:02,440 --> 00:29:04,760
Speaker 4: that brings my cost down. I sell more products, I

542
00:29:04,800 --> 00:29:07,680
Speaker 4: sell more service, I sell more anything. So what I

543
00:29:07,800 --> 00:29:09,720
Speaker 4: have to do to get there is I have to

544
00:29:09,720 --> 00:29:13,200
Speaker 4: figure out. Okay, what's my objective? I have a couple objectives. One,

545
00:29:13,280 --> 00:29:15,800
Speaker 4: I want customers to be able to support themselves without

546
00:29:15,800 --> 00:29:19,680
Speaker 4: even calling me, first off, and I don't want when

547
00:29:19,720 --> 00:29:22,120
Speaker 4: they call for the first answer to come back to

548
00:29:22,160 --> 00:29:25,520
Speaker 4: be did you try rebooting? Because I think that irritates

549
00:29:25,560 --> 00:29:27,960
Speaker 4: every single one of us. Did you try? Of course

550
00:29:27,960 --> 00:29:32,200
Speaker 4: I tried rebooting. I've had a laptop, of course I well, okay,

551
00:29:32,240 --> 00:29:35,200
Speaker 4: well then tell me, okay, what firmware version, all that

552
00:29:35,240 --> 00:29:38,840
Speaker 4: other stuff. Okay, we know this interaction. So that's kind

553
00:29:38,840 --> 00:29:40,720
Speaker 4: of the problem set. Do I want that to be

554
00:29:40,880 --> 00:29:44,840
Speaker 4: customers solving their own problems? Well, even for my support agents,

555
00:29:44,880 --> 00:29:46,920
Speaker 4: I want something in their pocket on their phone where

556
00:29:46,960 --> 00:29:49,520
Speaker 4: they say I'm seeing these symptoms and says, oh, this

557
00:29:49,680 --> 00:29:52,400
Speaker 4: happening around the globe. Here's kind of specific me. So

558
00:29:52,400 --> 00:29:56,000
Speaker 4: there's my problems. What does it mean for infrastructure on

559
00:29:56,040 --> 00:29:59,880
Speaker 4: the back end? So first I got to get all

560
00:29:59,880 --> 00:30:02,680
Speaker 4: that data together, right, all of those customer law, all

561
00:30:02,760 --> 00:30:05,960
Speaker 4: that customer support around the globe, et cetera. That needs

562
00:30:06,040 --> 00:30:08,560
Speaker 4: to be stored. That's a big set of data. And

563
00:30:08,640 --> 00:30:11,960
Speaker 4: some of it's not just fix and that kind of thing.

564
00:30:12,040 --> 00:30:14,480
Speaker 4: Some of it is Okay, you know what was the

565
00:30:14,480 --> 00:30:17,120
Speaker 4: firmware version? Who was the tech because it can matter.

566
00:30:17,640 --> 00:30:19,720
Speaker 4: Is this their first time fixing this problem? Is it

567
00:30:19,720 --> 00:30:22,040
Speaker 4: there one hundred and fiftieth time? What's their level? It's

568
00:30:22,040 --> 00:30:27,520
Speaker 4: a very complicated problem. Ingesting all that data takes an architecture.

569
00:30:27,520 --> 00:30:30,440
Speaker 4: We have a product called Scale, which is one of

570
00:30:30,440 --> 00:30:33,920
Speaker 4: our storage projects that actually makes it easy to ingest

571
00:30:33,920 --> 00:30:37,280
Speaker 4: all that data, get it organized, et cetera, and then

572
00:30:38,720 --> 00:30:41,320
Speaker 4: have a model. It's a whole different process to kind

573
00:30:41,320 --> 00:30:43,120
Speaker 4: of say did we train our model? We can train

574
00:30:43,160 --> 00:30:45,440
Speaker 4: our own models inside of IBM. We have a granite

575
00:30:45,440 --> 00:30:48,800
Speaker 4: set of models. Those models we fine tune, and then

576
00:30:48,840 --> 00:30:51,200
Speaker 4: we inference based on those models. So we can do

577
00:30:51,280 --> 00:30:54,080
Speaker 4: that inferencing in our cloud I have a cloud set

578
00:30:54,120 --> 00:30:56,640
Speaker 4: of infrastructure, or in my power servers. We can do

579
00:30:56,680 --> 00:31:01,520
Speaker 4: inferencing with our capabilities and say, okay, based on what

580
00:31:01,640 --> 00:31:04,360
Speaker 4: I'm saying, here's what the remediation that you should do

581
00:31:04,440 --> 00:31:07,720
Speaker 4: for that customer. We already are doing that today. We've

582
00:31:07,760 --> 00:31:13,360
Speaker 4: seen over a third of our support calls have had

583
00:31:13,520 --> 00:31:16,560
Speaker 4: significant reduction in the amount of time that it takes

584
00:31:16,640 --> 00:31:20,040
Speaker 4: to resolve that support call just by what I said

585
00:31:20,120 --> 00:31:20,560
Speaker 4: right there.

586
00:31:20,920 --> 00:31:24,479
Speaker 3: That I've really been curious about this. If I had

587
00:31:24,480 --> 00:31:28,719
Speaker 3: reduced something like AI into that equation as you just did. Yeah,

588
00:31:28,760 --> 00:31:31,760
Speaker 3: and you said we've already seen a thirty percent Say

589
00:31:31,800 --> 00:31:33,160
Speaker 3: did you say thirty percent reduction?

590
00:31:33,480 --> 00:31:38,040
Speaker 4: Thirty percent of our interactions have seen significant reduction in

591
00:31:38,440 --> 00:31:39,000
Speaker 4: those time?

592
00:31:39,160 --> 00:31:42,160
Speaker 3: Was that your primary goal to reduce the time of

593
00:31:42,200 --> 00:31:45,080
Speaker 3: the interaction? But it was you if everything else was

594
00:31:45,120 --> 00:31:47,719
Speaker 3: the same all, but what you were doing was shrinking

595
00:31:47,760 --> 00:31:48,400
Speaker 3: the amount of time?

596
00:31:48,440 --> 00:31:51,200
Speaker 4: That would you want one of the primary goals, So

597
00:31:52,320 --> 00:31:56,000
Speaker 4: to us in that business net promoter score kind of

598
00:31:56,000 --> 00:31:59,160
Speaker 4: the satisfaction of a client is the supreme goal. What

599
00:31:59,320 --> 00:32:02,720
Speaker 4: makes them sad fine, doesn't cost me a fortune, happens

600
00:32:02,760 --> 00:32:05,080
Speaker 4: really quickly, and if I can do it myself, I'd

601
00:32:05,080 --> 00:32:08,640
Speaker 4: be thrilled. It affects all of those right. It kind

602
00:32:08,640 --> 00:32:11,160
Speaker 4: of says it got resolved faster, it didn't cost me

603
00:32:11,200 --> 00:32:13,240
Speaker 4: an arm and the leg because the deck was barely here,

604
00:32:13,280 --> 00:32:16,320
Speaker 4: because it's a common problem, or I solved it myself

605
00:32:16,360 --> 00:32:20,320
Speaker 4: without even calling, So all of those objectives would kind

606
00:32:20,320 --> 00:32:22,280
Speaker 4: of hit across all so that now you see it.

607
00:32:22,320 --> 00:32:24,400
Speaker 4: So that's a little microcosm. That's just me and my

608
00:32:24,480 --> 00:32:27,640
Speaker 4: customer support business. Now think of how many problems for

609
00:32:27,800 --> 00:32:31,040
Speaker 4: businesses around the world there are like that. It's not

610
00:32:31,080 --> 00:32:34,479
Speaker 4: a it's not like a new AI application that changes

611
00:32:34,520 --> 00:32:39,280
Speaker 4: the entire user experience. That's those will come, But right

612
00:32:39,320 --> 00:32:42,120
Speaker 4: now it's kind of practical, which is, I just want

613
00:32:42,120 --> 00:32:45,000
Speaker 4: to do what I'm doing better and faster, and I

614
00:32:45,000 --> 00:32:47,680
Speaker 4: can get immediate economic return from those things.

615
00:32:47,680 --> 00:32:50,760
Speaker 3: How long How long did it take you to just

616
00:32:50,840 --> 00:32:54,200
Speaker 3: stick with that example of the customer reaction reducing thirty

617
00:32:54,200 --> 00:32:56,880
Speaker 3: percent of the time? How long from the very beginning

618
00:32:56,920 --> 00:33:00,240
Speaker 3: of that project, Yeah to that thirty percent reduction it was.

619
00:33:00,240 --> 00:33:05,440
Speaker 4: Hell long, less than a year. And yeah, So one

620
00:33:05,480 --> 00:33:08,440
Speaker 4: of the challenges, and this is interesting with a very

621
00:33:08,520 --> 00:33:12,360
Speaker 4: large organization, as you can imagine, just like you're seeing

622
00:33:12,360 --> 00:33:16,400
Speaker 4: in the industry, we don't have a problem of generating

623
00:33:16,520 --> 00:33:19,360
Speaker 4: ideas for how AI could help us. We actually have

624
00:33:19,480 --> 00:33:24,200
Speaker 4: a problem filtering the thousands of ideas from our employees

625
00:33:24,240 --> 00:33:26,720
Speaker 4: and from everywhere. It's like, hey, we could use AI

626
00:33:26,800 --> 00:33:29,160
Speaker 4: to and filtering down and saying, okay, which of these

627
00:33:29,200 --> 00:33:32,640
Speaker 4: will have a return on investment quickly and at a

628
00:33:32,720 --> 00:33:35,840
Speaker 4: level that sustains that's worth kind of going and investing

629
00:33:35,880 --> 00:33:39,600
Speaker 4: in the infrastructure and the software and kind of making

630
00:33:39,640 --> 00:33:41,040
Speaker 4: that happen. Is that unusual.

631
00:33:41,840 --> 00:33:44,520
Speaker 3: If I talked to you twenty five years ago and said,

632
00:33:44,920 --> 00:33:46,840
Speaker 3: do you have a problem of too many good ideas

633
00:33:46,920 --> 00:33:47,360
Speaker 3: or too few?

634
00:33:47,400 --> 00:33:53,560
Speaker 4: What was you said in this specific area, Probably too few,

635
00:33:53,720 --> 00:33:57,120
Speaker 4: because at some point you reach diminishing returns. So, for example,

636
00:33:57,160 --> 00:34:01,480
Speaker 4: let's use this same example. Can those thirteen thousand technicians

637
00:34:01,520 --> 00:34:05,880
Speaker 4: go faster? Can they spend less time driving to the side.

638
00:34:05,880 --> 00:34:07,600
Speaker 4: I mean, there's only so much you can kind of

639
00:34:07,640 --> 00:34:10,000
Speaker 4: do on those things. But if you can get them

640
00:34:10,000 --> 00:34:12,279
Speaker 4: an answer to the problem and maybe even avoid them

641
00:34:12,320 --> 00:34:14,880
Speaker 4: having to visit at all because the client helped themselves,

642
00:34:15,280 --> 00:34:18,279
Speaker 4: that's a step function. So that's why people are kind

643
00:34:18,280 --> 00:34:22,680
Speaker 4: of talking about there's a business revolution coming with AI

644
00:34:22,800 --> 00:34:25,799
Speaker 4: where there are some step function changes that can be there.

645
00:34:25,840 --> 00:34:29,160
Speaker 4: And notice I didn't say I'm going to have less

646
00:34:29,200 --> 00:34:32,799
Speaker 4: of those agents. That's not my objective. My objective and

647
00:34:32,840 --> 00:34:35,200
Speaker 4: I think that's the fear in the industry about AI

648
00:34:35,280 --> 00:34:37,640
Speaker 4: is going to eliminate all the jobs. No, I just

649
00:34:37,719 --> 00:34:41,680
Speaker 4: created thirteen thousand superpowered agents that can do more right.

650
00:34:41,760 --> 00:34:43,960
Speaker 4: And so I'm not just going to support IBM products.

651
00:34:44,200 --> 00:34:46,360
Speaker 4: I'm going to go out and support other people's products

652
00:34:46,400 --> 00:34:48,040
Speaker 4: because I know how to do that really well. And

653
00:34:48,080 --> 00:34:50,840
Speaker 4: once I have the data on how to fix their problems,

654
00:34:51,280 --> 00:34:54,920
Speaker 4: I may just have a customer support business that's independent

655
00:34:54,960 --> 00:34:57,680
Speaker 4: of my boxes. So you know, I think that's where

656
00:34:57,719 --> 00:35:00,279
Speaker 4: people sometimes get it wrong. And the AI thing is,

657
00:35:00,840 --> 00:35:04,799
Speaker 4: it's like, you know, did word processing eliminate the need

658
00:35:04,880 --> 00:35:10,280
Speaker 4: for writers? No? It enabled writing instead of mucking around

659
00:35:10,280 --> 00:35:13,239
Speaker 4: with mimeographic machines and click and click typewriters. It may

660
00:35:13,280 --> 00:35:16,400
Speaker 4: have enabled too much writing? Yeah, maybe maybe can I

661
00:35:16,400 --> 00:35:17,440
Speaker 4: give you a hypothetical?

662
00:35:18,520 --> 00:35:18,759
Speaker 2: Uh?

663
00:35:18,840 --> 00:35:20,759
Speaker 3: And I asked this because I ran I was at

664
00:35:20,800 --> 00:35:22,800
Speaker 3: some conmis and I ran into some guy from the

665
00:35:22,960 --> 00:35:26,200
Speaker 3: I R S who was really, really, really really excited

666
00:35:26,200 --> 00:35:31,440
Speaker 3: about AI. So let's suppose they call you up and

667
00:35:31,520 --> 00:35:35,920
Speaker 3: they say you're going to talk to the I R SKY.

668
00:35:35,960 --> 00:35:40,680
Speaker 3: I call you up and I say, Rick, Uh, clearly

669
00:35:40,719 --> 00:35:44,279
Speaker 3: there's something that we could do for the I R

670
00:35:44,400 --> 00:35:45,440
Speaker 3: S if we work together.

671
00:35:45,640 --> 00:35:48,480
Speaker 4: Yeah, what would your answer? Of course?

672
00:35:49,160 --> 00:35:49,239
Speaker 2: No.

673
00:35:49,360 --> 00:35:53,080
Speaker 4: I think we sell to a lot of government agencies.

674
00:35:53,160 --> 00:35:56,719
Speaker 4: I can imagine in the business that we're in, we

675
00:35:56,880 --> 00:36:00,520
Speaker 4: enable a lot of social security transaction and things like

676
00:36:00,560 --> 00:36:04,799
Speaker 4: that through our mainframes. And I think, you know, we're

677
00:36:04,800 --> 00:36:08,200
Speaker 4: in the business of helping whatever client get the most

678
00:36:08,239 --> 00:36:10,319
Speaker 4: out of their data and be able to secure it

679
00:36:10,360 --> 00:36:13,880
Speaker 4: and be able to do analytics with this. And IRS

680
00:36:13,880 --> 00:36:16,040
Speaker 4: has a heck of a lot of data, so yes,

681
00:36:16,080 --> 00:36:16,839
Speaker 4: we would help them.

682
00:36:17,200 --> 00:36:18,920
Speaker 3: Do you know how the amount of data they have

683
00:36:19,040 --> 00:36:21,319
Speaker 3: compares to some of the corporate clients you've I.

684
00:36:21,280 --> 00:36:24,400
Speaker 4: Don't know specifically for the IRS how much data they have,

685
00:36:24,480 --> 00:36:27,080
Speaker 4: but I would assume it's a whole lot. It's mountains.

686
00:36:27,120 --> 00:36:31,359
Speaker 4: But that's our business. I mean, it's interesting sometimes people

687
00:36:31,400 --> 00:36:34,319
Speaker 4: of that what's the most you know, what is it

688
00:36:34,440 --> 00:36:39,520
Speaker 4: that IBM has that's of great value? Is it a server?

689
00:36:39,800 --> 00:36:42,960
Speaker 4: Is it a storage array? Is it you know, software

690
00:36:42,960 --> 00:36:46,520
Speaker 4: and all that. What we have is the most important

691
00:36:46,719 --> 00:36:50,400
Speaker 4: entities in the world have their data on our stuff.

692
00:36:50,520 --> 00:36:54,279
Speaker 4: The most important data in the world. It's not you know,

693
00:36:54,560 --> 00:36:57,520
Speaker 4: pictures of your grandkids and things like that. Generally for us,

694
00:36:57,600 --> 00:37:00,680
Speaker 4: it's all of the financial transactions that have and globally,

695
00:37:00,760 --> 00:37:03,640
Speaker 4: right it's all of the it's the world's economy is

696
00:37:03,719 --> 00:37:07,080
Speaker 4: kind of running through our systems, and so we take

697
00:37:07,120 --> 00:37:10,319
Speaker 4: that really seriously. You know, you would be distraught if

698
00:37:10,360 --> 00:37:13,160
Speaker 4: you lost one photo on your laptop or whatever, but

699
00:37:13,520 --> 00:37:16,440
Speaker 4: you know, if we lose a transaction, like somebody moves

700
00:37:16,440 --> 00:37:18,959
Speaker 4: a big amount of money and it's like, well, don't

701
00:37:18,960 --> 00:37:22,040
Speaker 4: know what happened there. It is a massive deal, right,

702
00:37:22,120 --> 00:37:23,600
Speaker 4: so that doesn't happen.

703
00:37:23,719 --> 00:37:25,240
Speaker 3: But I want to go back to my irs example

704
00:37:25,280 --> 00:37:29,160
Speaker 3: for US, Yes, so one, is it reasonable to assume

705
00:37:29,440 --> 00:37:34,480
Speaker 3: that you could that somebody IBM or somebody else could

706
00:37:34,520 --> 00:37:36,960
Speaker 3: in a short period of time put together not just

707
00:37:37,040 --> 00:37:42,600
Speaker 3: the AI capability to audit returns, but also this the

708
00:37:42,640 --> 00:37:45,879
Speaker 3: infrastructure support for that in a reasonable amount of time

709
00:37:45,880 --> 00:37:48,319
Speaker 3: for a reasonable amount of cost. Or is it over?

710
00:37:48,719 --> 00:37:51,560
Speaker 3: Is it going to the moon? Or is it it?

711
00:37:51,800 --> 00:37:55,640
Speaker 4: Definitely? I mean so we're already doing that kind of

712
00:37:55,680 --> 00:37:59,640
Speaker 4: thing right across a network of banks and others. Yeah,

713
00:38:00,440 --> 00:38:04,800
Speaker 4: essentially all credit card transactions for all of the world

714
00:38:04,880 --> 00:38:08,120
Speaker 4: to go through our systems, So that in some ways

715
00:38:08,160 --> 00:38:11,800
Speaker 4: is more volume than the datch returns of the US people.

716
00:38:11,920 --> 00:38:15,120
Speaker 4: And they're W two's and all that stuff, and we

717
00:38:15,200 --> 00:38:18,000
Speaker 4: do that stuff too. I try not to describe it

718
00:38:18,040 --> 00:38:20,239
Speaker 4: too much in detail, but we definitely do a lot

719
00:38:20,280 --> 00:38:25,920
Speaker 4: of that. In fact, I think most of if you think, okay,

720
00:38:25,920 --> 00:38:29,080
Speaker 4: what is super critical data, who would be doing the

721
00:38:29,120 --> 00:38:33,160
Speaker 4: business transaction processing it is most likely us in almost

722
00:38:33,239 --> 00:38:37,600
Speaker 4: all cases, whether it's government things or private or banks

723
00:38:37,920 --> 00:38:40,040
Speaker 4: or that kind of thing. That's what we do.

724
00:38:40,360 --> 00:38:42,360
Speaker 3: Rick we're going to end with the where we always

725
00:38:42,480 --> 00:38:45,080
Speaker 3: end with a couple of quick fire questions. Okay, here

726
00:38:45,120 --> 00:38:48,600
Speaker 3: we go. What single piece of advice would you give

727
00:38:48,719 --> 00:38:51,960
Speaker 3: to businesses trying to use AI in an effective way?

728
00:38:52,200 --> 00:38:55,759
Speaker 4: The simple version is get started. By get started, I

729
00:38:55,800 --> 00:38:59,840
Speaker 4: mean think of what is something that I want to improve.

730
00:39:00,000 --> 00:39:02,120
Speaker 4: The things that we have traction on right now in

731
00:39:02,160 --> 00:39:08,080
Speaker 4: the market are around business process, automation, digital labor, those

732
00:39:08,200 --> 00:39:11,400
Speaker 4: kind of things. But my other little piece of advice

733
00:39:11,440 --> 00:39:13,440
Speaker 4: there is keep it simple to begin with. You're going

734
00:39:13,520 --> 00:39:16,200
Speaker 4: to learn a lot, but getting started means you'll start

735
00:39:16,239 --> 00:39:20,040
Speaker 4: that learning curve. I even advise you My friends like, hey,

736
00:39:20,040 --> 00:39:22,600
Speaker 4: should I be playing around with some of this AI stuff?

737
00:39:22,640 --> 00:39:25,080
Speaker 4: And I say yeah, because I think it will help

738
00:39:25,120 --> 00:39:27,799
Speaker 4: you start to be more comfortable and you may find

739
00:39:27,840 --> 00:39:29,839
Speaker 4: a use case personally for that. I think the same

740
00:39:29,960 --> 00:39:33,000
Speaker 4: is true for businesses. The first step in that journey

741
00:39:33,040 --> 00:39:36,760
Speaker 4: is always with what data. Notice when I talked about

742
00:39:36,760 --> 00:39:41,080
Speaker 4: our customer support people, I thought about, Okay, what's the data.

743
00:39:41,239 --> 00:39:43,799
Speaker 4: The data is all of those logs of all of

744
00:39:43,800 --> 00:39:47,000
Speaker 4: those service engagements around the world, and what could I

745
00:39:47,040 --> 00:39:48,839
Speaker 4: do with that? Well, I could use that to get

746
00:39:48,880 --> 00:39:52,680
Speaker 4: to a knowledge base that really helps and hopefully that

747
00:39:52,719 --> 00:39:55,320
Speaker 4: I can do it in multiple languages because it's global

748
00:39:55,320 --> 00:39:57,840
Speaker 4: and I can you know, all of those things. That

749
00:39:57,960 --> 00:40:00,840
Speaker 4: was kind of my data sent That one's not super simple,

750
00:40:00,880 --> 00:40:03,640
Speaker 4: but we've had a lot of experience in AI for

751
00:40:03,719 --> 00:40:06,600
Speaker 4: other people that might just be how do I automate

752
00:40:06,719 --> 00:40:10,680
Speaker 4: filling out travel expense reports for my company? We can

753
00:40:10,680 --> 00:40:13,120
Speaker 4: help people that we have consulting, we have wats and

754
00:40:13,280 --> 00:40:15,399
Speaker 4: X tools. We can do that like this, and we're

755
00:40:15,440 --> 00:40:18,560
Speaker 4: doing it globally for people around the world. Pick that thing.

756
00:40:18,640 --> 00:40:21,600
Speaker 4: What's the data you have? In that case, it's data

757
00:40:21,600 --> 00:40:23,919
Speaker 4: of expense reports and it's like, okay, we can help

758
00:40:23,920 --> 00:40:26,200
Speaker 4: you automate that for people where they could do it

759
00:40:26,320 --> 00:40:30,120
Speaker 4: just by you know, a verbal interface. What did you spend,

760
00:40:30,200 --> 00:40:32,200
Speaker 4: where did you go? Who you were you with? Okay,

761
00:40:32,280 --> 00:40:34,560
Speaker 4: we filled out your travel expense report for you and

762
00:40:34,600 --> 00:40:35,880
Speaker 4: you don't have to mess around with it.

763
00:40:36,040 --> 00:40:38,600
Speaker 3: So we were playing with this idea where we would

764
00:40:39,000 --> 00:40:42,360
Speaker 3: pick a business and go in there and do it

765
00:40:42,400 --> 00:40:45,600
Speaker 3: would be AI makeover. Yeah, I love that what's okay,

766
00:40:45,640 --> 00:40:49,200
Speaker 3: what is the ideal business to do? We only have

767
00:40:49,239 --> 00:40:51,520
Speaker 3: a couple months. We don't want to spend a kajillion dollars.

768
00:40:51,760 --> 00:40:54,400
Speaker 3: We want to be able to show tangibly and quickly

769
00:40:54,480 --> 00:40:57,440
Speaker 3: what AI can do. What's an ideal business to do

770
00:40:57,480 --> 00:40:59,040
Speaker 3: that in It can be a small business. We're not

771
00:40:59,080 --> 00:41:01,400
Speaker 3: talking this grand corporate thing there.

772
00:41:02,120 --> 00:41:06,239
Speaker 4: Ah boy, small business that we could do and hey,

773
00:41:06,320 --> 00:41:10,560
Speaker 4: I make over. Customer support is one of my favorites

774
00:41:10,600 --> 00:41:13,440
Speaker 4: because it's a it's it's I have it on the

775
00:41:13,480 --> 00:41:16,840
Speaker 4: business side where I provide customer support. I have it

776
00:41:16,880 --> 00:41:19,839
Speaker 4: on the consumer side, where it drives me nuts when

777
00:41:19,840 --> 00:41:22,760
Speaker 4: I have to go through thirty layers of phone menus.

778
00:41:23,280 --> 00:41:25,319
Speaker 4: Speak to an agent, speak to an agent, speak to

779
00:41:25,360 --> 00:41:29,560
Speaker 4: an agent. That for any business, I think is just

780
00:41:29,800 --> 00:41:32,040
Speaker 4: ripe to be able to kind of say why do

781
00:41:32,120 --> 00:41:34,400
Speaker 4: I have to click through these manucent messages. I just

782
00:41:34,440 --> 00:41:37,040
Speaker 4: need to tell you in human language, here's the issue,

783
00:41:37,040 --> 00:41:39,800
Speaker 4: and I'll be really good about telling you details about

784
00:41:40,320 --> 00:41:42,759
Speaker 4: You know, I tried to set up this thing for

785
00:41:42,920 --> 00:41:44,680
Speaker 4: my bank and I do da da da da da.

786
00:41:44,960 --> 00:41:48,640
Speaker 4: They can go through all the menus automate that process.

787
00:41:48,920 --> 00:41:51,239
Speaker 4: I think it would change everything because all that frustration

788
00:41:51,360 --> 00:41:54,799
Speaker 4: as a consumer would go down dramatically, and it's all,

789
00:41:55,320 --> 00:41:58,320
Speaker 4: you know, why are you making me the beep booth

790
00:41:58,520 --> 00:42:03,840
Speaker 4: press one? Exactly? Well, don't offload to me, offload to AI.

791
00:42:04,320 --> 00:42:05,319
Speaker 4: We can help you with that.

792
00:42:05,600 --> 00:42:08,880
Speaker 3: Here's my version of that drives me crazy. Every morning

793
00:42:09,040 --> 00:42:12,080
Speaker 3: I go to the same coffee shop and I get

794
00:42:12,680 --> 00:42:14,800
Speaker 3: a cup of tea and a croissant.

795
00:42:15,160 --> 00:42:16,080
Speaker 4: And here's what happens.

796
00:42:16,080 --> 00:42:18,799
Speaker 3: A person has their screen and they go, I go,

797
00:42:19,080 --> 00:42:27,319
Speaker 3: cup of tea, croissant, sparkling water, like at least twenty keystrokes,

798
00:42:28,080 --> 00:42:30,719
Speaker 3: and then like then the screen is turned around. Like

799
00:42:30,800 --> 00:42:33,160
Speaker 3: at this point we're like forty five seconds in, I'm like,

800
00:42:33,440 --> 00:42:35,359
Speaker 3: why is this? First of all, it's not for me,

801
00:42:35,440 --> 00:42:39,000
Speaker 3: all those keystrokes, it's their internal right, right, So they're

802
00:42:39,000 --> 00:42:40,400
Speaker 3: burdening me in order to service.

803
00:42:40,440 --> 00:42:42,160
Speaker 4: To back it, you should be able to walk in,

804
00:42:42,400 --> 00:42:44,399
Speaker 4: go up and they go, I'm olc them the same

805
00:42:44,440 --> 00:42:46,600
Speaker 4: thing and you just go yes, and then.

806
00:42:46,560 --> 00:42:48,880
Speaker 3: The boom, We're done. Can we do AI makeover of

807
00:42:48,960 --> 00:42:49,680
Speaker 3: my coffee shop?

808
00:42:51,560 --> 00:42:55,239
Speaker 4: You notice I quickly jumped more to banks than your

809
00:42:55,320 --> 00:42:58,239
Speaker 4: coffee shop because I think I'm a business person, but

810
00:42:58,600 --> 00:43:00,759
Speaker 4: I'm not trying to kind of do a deal on

811
00:43:00,760 --> 00:43:01,600
Speaker 4: one coffee shop.

812
00:43:01,920 --> 00:43:03,880
Speaker 3: But this is interesting because it takes me back to

813
00:43:03,960 --> 00:43:06,879
Speaker 3: something you said that I thought was really important. When

814
00:43:06,920 --> 00:43:09,680
Speaker 3: you were talking about when you were using AI and

815
00:43:09,719 --> 00:43:12,880
Speaker 3: your customer service thing, it was clear that your goal

816
00:43:13,040 --> 00:43:15,560
Speaker 3: you could have any number of goals, yes, going in.

817
00:43:15,760 --> 00:43:18,840
Speaker 3: It could be to cut costs, it could be to

818
00:43:18,920 --> 00:43:20,520
Speaker 3: dramatically improved profits.

819
00:43:21,080 --> 00:43:21,920
Speaker 4: Your goal, quite.

820
00:43:21,760 --> 00:43:24,640
Speaker 3: Specifically, was to improve the experience of your customer, right,

821
00:43:24,680 --> 00:43:25,160
Speaker 3: So you were.

822
00:43:25,120 --> 00:43:27,879
Speaker 4: Using it to that. All the other things come from

823
00:43:27,960 --> 00:43:31,560
Speaker 4: that come from. That is actually one of the beautiful

824
00:43:31,600 --> 00:43:35,440
Speaker 4: pillars of the IBM culture is delighting clients is actually

825
00:43:35,480 --> 00:43:37,640
Speaker 4: where all of the good stuff comes from.

826
00:43:37,680 --> 00:43:41,680
Speaker 3: So my coffee shop thing is the same principle. Right now,

827
00:43:41,960 --> 00:43:45,000
Speaker 3: they're making my customer experience worse and they don't want to,

828
00:43:45,480 --> 00:43:48,600
Speaker 3: but their eyes are glued to the special a moment

829
00:43:48,640 --> 00:43:50,440
Speaker 3: when I walk in and I want to say, Hi,

830
00:43:50,560 --> 00:43:53,799
Speaker 3: how are you doing? We could have a conversation. You're

831
00:43:53,800 --> 00:43:56,960
Speaker 3: too busy, busy pooping, So like, this is the same thing.

832
00:43:57,040 --> 00:43:58,799
Speaker 3: If they had it that, oh, this isn't if they

833
00:43:58,920 --> 00:44:01,520
Speaker 3: understood they had an operation need to improve the experience

834
00:44:01,520 --> 00:44:03,120
Speaker 3: of their customer experience.

835
00:44:02,800 --> 00:44:07,040
Speaker 4: I would not be surprised if a chain comes along

836
00:44:07,480 --> 00:44:09,960
Speaker 4: where that is their value proposition. I would not be

837
00:44:10,000 --> 00:44:14,439
Speaker 4: surprised at all. Yeah, yeah, right, So I mean and

838
00:44:14,440 --> 00:44:17,720
Speaker 4: and when those things kind of catch hold, it becomes

839
00:44:17,719 --> 00:44:18,360
Speaker 4: a revolution.

840
00:44:18,600 --> 00:44:20,560
Speaker 3: You know, when the guy comes to do like to

841
00:44:20,560 --> 00:44:23,240
Speaker 3: redo your roof and they put a sign out front,

842
00:44:23,320 --> 00:44:25,920
Speaker 3: like you know, Joe's roofing. You guys could do the

843
00:44:25,920 --> 00:44:29,640
Speaker 3: same with my coffee shop. But like I'd be i'ure

844
00:44:29,840 --> 00:44:33,280
Speaker 3: was here exactly exactly.

845
00:44:35,200 --> 00:44:39,160
Speaker 4: In five years, the main frame will be dot dot

846
00:44:39,200 --> 00:44:46,920
Speaker 4: dot going strong, the mainframe going strong and with new capabilities,

847
00:44:47,000 --> 00:44:51,560
Speaker 4: continuous new capabilities. I think when we announced the last

848
00:44:51,680 --> 00:44:55,440
Speaker 4: version Z sixteen, the latest version, I should say, and

849
00:44:55,480 --> 00:44:58,880
Speaker 4: we said, hey, there's AI processing built into it. This

850
00:44:59,080 --> 00:45:01,480
Speaker 4: was before everybody was talking about that. I think a

851
00:45:01,480 --> 00:45:04,000
Speaker 4: lot of people thought, what's that for? And we did

852
00:45:04,040 --> 00:45:07,920
Speaker 4: it specifically for traditional AI fraud detection, et cetera. This

853
00:45:08,080 --> 00:45:10,759
Speaker 4: next version, not only do we have the traditional AI

854
00:45:10,880 --> 00:45:13,759
Speaker 4: built in, but we have optional cards that you can

855
00:45:13,760 --> 00:45:16,520
Speaker 4: plug into it to allow you to do large language

856
00:45:16,560 --> 00:45:21,720
Speaker 4: models for the enhanced fraud detection cases that we talked about,

857
00:45:21,840 --> 00:45:25,720
Speaker 4: where you know, it's more than just what transactions were happening.

858
00:45:25,840 --> 00:45:29,319
Speaker 4: So if you take that and say, okay, the next generations,

859
00:45:30,360 --> 00:45:34,040
Speaker 4: we have more transaction volume than we've ever had in mainframes. Today,

860
00:45:34,280 --> 00:45:38,080
Speaker 4: the business is growing, it's strong, we keep innovating. In

861
00:45:38,120 --> 00:45:39,680
Speaker 4: five years it'll be going strong.

862
00:45:39,800 --> 00:45:42,400
Speaker 3: But we're people. You're saying this in the context of

863
00:45:43,280 --> 00:45:45,480
Speaker 3: for years people were predicting, weren't they that the main

864
00:45:45,520 --> 00:45:46,560
Speaker 3: brand was going to go away.

865
00:45:48,440 --> 00:45:50,879
Speaker 4: There were pundits in the market that said everything will

866
00:45:50,880 --> 00:45:52,840
Speaker 4: go away there, no one will ever have a box,

867
00:45:52,880 --> 00:45:55,560
Speaker 4: It'll all be online. I think this is something I've

868
00:45:55,640 --> 00:46:00,200
Speaker 4: learned big time in my long career. You know in

869
00:46:00,239 --> 00:46:04,440
Speaker 4: the IT industry is don't believe everything you hear. So

870
00:46:04,520 --> 00:46:08,600
Speaker 4: I went back for my master's degree at Stanford after

871
00:46:08,640 --> 00:46:13,240
Speaker 4: I had worked a while in as a hardware designer,

872
00:46:13,520 --> 00:46:16,360
Speaker 4: and everybody told me be sure to do your masters

873
00:46:16,360 --> 00:46:19,400
Speaker 4: in software. Hardware is dead. I went on to work

874
00:46:19,640 --> 00:46:23,200
Speaker 4: for thirty plus years in hardware and infrastructure. Now software

875
00:46:23,200 --> 00:46:25,640
Speaker 4: became important, and I'm glad I had that extra training

876
00:46:25,640 --> 00:46:28,440
Speaker 4: in software because it helped me in hardware. But hardware

877
00:46:28,520 --> 00:46:32,080
Speaker 4: wasn't dead. Then I heard all infrastructure will go into

878
00:46:32,120 --> 00:46:35,160
Speaker 4: the cloud. There won't be that hasn't happened, it's not happening.

879
00:46:35,440 --> 00:46:37,880
Speaker 4: Then I heard there will only be one cloud because

880
00:46:37,920 --> 00:46:40,360
Speaker 4: one of the players will dominate. There's not one cloud.

881
00:46:40,440 --> 00:46:44,680
Speaker 4: So I think it's as humans we like to oversimplify

882
00:46:44,719 --> 00:46:46,960
Speaker 4: and go, oh, it's all going to be this, And

883
00:46:47,080 --> 00:46:51,160
Speaker 4: kind of what I've learned is fit for purpose matters

884
00:46:51,200 --> 00:46:56,960
Speaker 4: in everything. It matters in size of infrastructure, it matters

885
00:46:56,960 --> 00:46:59,640
Speaker 4: in the stack that goes along with solving a specific

886
00:46:59,760 --> 00:47:03,040
Speaker 4: use case. If you're willing to design something that's the

887
00:47:03,040 --> 00:47:05,360
Speaker 4: best at that use case, if you're willing to design

888
00:47:05,440 --> 00:47:07,759
Speaker 4: the coffee shop that is the best at greeting me,

889
00:47:08,120 --> 00:47:10,000
Speaker 4: there's a spot for you, and there may be a

890
00:47:10,040 --> 00:47:13,920
Speaker 4: big business in doing that. So oversimplifying is really.

891
00:47:13,920 --> 00:47:17,120
Speaker 3: When you heard all those predictions, did you believe them

892
00:47:17,120 --> 00:47:17,920
Speaker 3: at the time.

893
00:47:19,280 --> 00:47:22,040
Speaker 4: They looked like they were trending in that direction. I'll

894
00:47:22,080 --> 00:47:24,759
Speaker 4: tell you some right now which might be useful. There

895
00:47:24,760 --> 00:47:27,160
Speaker 4: will only be one GPU company and they're going to

896
00:47:27,880 --> 00:47:30,480
Speaker 4: end up taking over the world. It's a pretty obvious answer.

897
00:47:30,520 --> 00:47:33,920
Speaker 4: Whose economic values risen dramatically. I don't think that's going

898
00:47:33,960 --> 00:47:36,400
Speaker 4: to be the case. In fact, I think that ninety

899
00:47:36,480 --> 00:47:42,080
Speaker 4: percent of processing for AI actually happen happens at inferencing,

900
00:47:42,520 --> 00:47:46,000
Speaker 4: and inferencing is not as GPU and hardware intensive as

901
00:47:46,040 --> 00:47:48,480
Speaker 4: the other things and is a lot more amenable to

902
00:47:48,600 --> 00:47:51,680
Speaker 4: fit for purpose. So the model size will matter. The

903
00:47:51,800 --> 00:47:54,120
Speaker 4: tuning matters a lot. As we're learning. We have a

904
00:47:54,120 --> 00:47:58,200
Speaker 4: product around instruct lab that's really focused on tuning. So

905
00:47:58,640 --> 00:48:00,640
Speaker 4: that was one thing is there'll be one GPU. The

906
00:48:00,640 --> 00:48:04,560
Speaker 4: other thing is that the biggest model will win. I

907
00:48:04,600 --> 00:48:06,959
Speaker 4: think is another thing that's kind of people are saying

908
00:48:07,040 --> 00:48:09,160
Speaker 4: right now. Don't believe that I believe they will be

909
00:48:09,239 --> 00:48:12,520
Speaker 4: fit for purpose models. It takes a lot of money

910
00:48:12,520 --> 00:48:15,840
Speaker 4: to run to create a huge model, and then to

911
00:48:16,000 --> 00:48:18,719
Speaker 4: run a huge model, or to even infer off of

912
00:48:18,760 --> 00:48:22,080
Speaker 4: a huge model. I don't need a massive training GPU

913
00:48:22,280 --> 00:48:26,240
Speaker 4: set thing to solve my thirteen thousand people customer support issues.

914
00:48:26,280 --> 00:48:28,399
Speaker 4: So why would I feel like I got to go

915
00:48:28,520 --> 00:48:30,920
Speaker 4: farm that out for a big expensive thing. I can

916
00:48:30,960 --> 00:48:32,920
Speaker 4: do that on a small box. In some cases I

917
00:48:33,000 --> 00:48:34,719
Speaker 4: might even be able to do that on a laptop.

918
00:48:35,239 --> 00:48:37,080
Speaker 4: The other thing I'll say in this we are so

919
00:48:37,320 --> 00:48:40,000
Speaker 4: early innings in AI, A lot of things are going

920
00:48:40,080 --> 00:48:42,480
Speaker 4: to change. So anybody kind of saying it will all

921
00:48:42,520 --> 00:48:44,960
Speaker 4: be X, Y or Z, I just think you have

922
00:48:45,080 --> 00:48:47,000
Speaker 4: no idea how this is going to play out, and

923
00:48:47,680 --> 00:48:49,360
Speaker 4: it's up to us to go figure out how it

924
00:48:49,360 --> 00:48:49,879
Speaker 4: plays out.

925
00:48:50,160 --> 00:48:54,040
Speaker 3: Yeah, yeah, all right, in five years AI will be

926
00:48:54,400 --> 00:48:56,560
Speaker 3: dot dot dot still new.

927
00:48:59,040 --> 00:49:03,120
Speaker 4: It will have moved a bunch in five years, but

928
00:49:03,160 --> 00:49:07,520
Speaker 4: the potential for the disruption in the world will still

929
00:49:07,600 --> 00:49:10,600
Speaker 4: will still be very early innings in that process. And

930
00:49:10,640 --> 00:49:13,000
Speaker 4: I think that's super important to realize. That's why I

931
00:49:13,040 --> 00:49:16,439
Speaker 4: say get started, start thinking about how that could change,

932
00:49:16,440 --> 00:49:19,279
Speaker 4: because it'll be some little things first, but it will

933
00:49:19,320 --> 00:49:20,400
Speaker 4: continue to snowball.

934
00:49:20,600 --> 00:49:26,080
Speaker 3: This is a common observation that we the invention of

935
00:49:26,120 --> 00:49:31,319
Speaker 3: the capability massively predates the understanding.

936
00:49:30,760 --> 00:49:33,239
Speaker 4: Of the capability, right, Like I love that.

937
00:49:33,440 --> 00:49:39,279
Speaker 3: Yeah, Like yes, recorded recording shows on television is invented

938
00:49:39,360 --> 00:49:45,480
Speaker 3: in the sixties. Probably we don't really understand what it's

939
00:49:45,560 --> 00:49:50,359
Speaker 3: used for until the oughts was what's really good for

940
00:49:50,480 --> 00:49:54,000
Speaker 3: is being able to tell a story sequentially, Yes, over time,

941
00:49:54,080 --> 00:49:55,960
Speaker 3: because you know that the person will always see in

942
00:49:56,000 --> 00:49:58,760
Speaker 3: the episode before, so you get the Sopranos. And yes, yes,

943
00:49:58,840 --> 00:50:03,000
Speaker 3: Hollywood wanted to ban the VCR in the beginning, Yeah,

944
00:50:03,160 --> 00:50:04,880
Speaker 3: because they thought it was good. They thought the point

945
00:50:04,880 --> 00:50:08,560
Speaker 3: of it was thought then understand. No, No, No, it's storytelling.

946
00:50:08,719 --> 00:50:11,600
Speaker 3: It's actually your business is getting better. Yes, took them

947
00:50:11,640 --> 00:50:13,920
Speaker 3: twenty years to figure that out, which is to your point,

948
00:50:14,320 --> 00:50:16,279
Speaker 3: why would we know what AI was four and five years.

949
00:50:16,520 --> 00:50:18,600
Speaker 4: Well, that's why you hear people kind of say, oh

950
00:50:18,680 --> 00:50:21,680
Speaker 4: my gosh, AI's that will just eliminate jobs. No, it'll

951
00:50:21,680 --> 00:50:23,520
Speaker 4: make jobs better. That's how I view it.

952
00:50:23,640 --> 00:50:27,120
Speaker 3: Yeah, what's the number one thing that people misunderstand about AI?

953
00:50:27,280 --> 00:50:30,520
Speaker 4: Is that it that it'll I think that's that. That

954
00:50:30,560 --> 00:50:33,480
Speaker 4: would be the human kind of understanding part of it.

955
00:50:33,560 --> 00:50:36,880
Speaker 4: The technology part of it, I think would be what

956
00:50:37,040 --> 00:50:41,239
Speaker 4: I was talking about fit for purpose, meaning that it

957
00:50:41,280 --> 00:50:44,080
Speaker 4: isn't just going to be a GPU arms race all

958
00:50:44,120 --> 00:50:46,680
Speaker 4: of AI. I don't believe that at all. It will

959
00:50:46,760 --> 00:50:49,080
Speaker 4: change everything, but it's not just going to be a GPU.

960
00:50:48,880 --> 00:50:52,480
Speaker 3: Armed Next question, what advice would you give yourself ten

961
00:50:52,560 --> 00:50:54,520
Speaker 3: years ago to better prepare you for today?

962
00:50:54,920 --> 00:50:56,800
Speaker 4: I'm changing this question, okay.

963
00:50:58,000 --> 00:51:02,000
Speaker 3: I would say, let's imagine that at what was your

964
00:51:02,040 --> 00:51:03,120
Speaker 3: what college.

965
00:51:02,840 --> 00:51:06,120
Speaker 4: You to go to? I went to three of them.

966
00:51:06,160 --> 00:51:09,520
Speaker 4: My undergrad was Utah State University, my NBA was Santa

967
00:51:09,520 --> 00:51:13,160
Speaker 4: Clara University, and my master's in w was Stanford.

968
00:51:12,840 --> 00:51:16,080
Speaker 3: Okay, any one of those three culture up and says

969
00:51:16,440 --> 00:51:21,280
Speaker 3: we want you to give the commencement address and imagine

970
00:51:21,320 --> 00:51:24,440
Speaker 3: it's it's it's let's just say, for the sake of argument,

971
00:51:24,480 --> 00:51:25,960
Speaker 3: it's just to the stem people.

972
00:51:26,440 --> 00:51:30,600
Speaker 4: Those are the relevant parties here. What do you tell them? Boy?

973
00:51:30,680 --> 00:51:36,759
Speaker 4: What do I tell them? Let's see. I think I

974
00:51:36,800 --> 00:51:41,400
Speaker 4: would start with life is a marathon, not a sprint.

975
00:51:41,800 --> 00:51:45,640
Speaker 4: It would be the first one. The second thing I

976
00:51:45,640 --> 00:51:48,920
Speaker 4: would say that in that spirit is be sure to

977
00:51:49,040 --> 00:51:54,200
Speaker 4: set yourself some big, hairy, audacious goals and don't be

978
00:51:54,400 --> 00:51:59,719
Speaker 4: overly disappointed if you don't hit them all. Going after

979
00:51:59,760 --> 00:52:02,480
Speaker 4: those big, hairy, audacious goals will get you on a

980
00:52:02,520 --> 00:52:06,720
Speaker 4: path where you will learn so much. You will achieve

981
00:52:06,800 --> 00:52:09,520
Speaker 4: more than you ever could imagine you would have achieved.

982
00:52:09,719 --> 00:52:11,920
Speaker 4: That's what the advice I give to my kids is, ye,

983
00:52:12,160 --> 00:52:15,040
Speaker 4: set some big goals, get after it. You may or

984
00:52:15,040 --> 00:52:16,759
Speaker 4: may not achieve them, but you'll be better for the

985
00:52:16,760 --> 00:52:17,880
Speaker 4: whole process when you're done.

986
00:52:17,880 --> 00:52:20,280
Speaker 3: By the way, as someone whose kids are younger than yours,

987
00:52:21,080 --> 00:52:23,360
Speaker 3: is it actually useful to give your give advice to

988
00:52:23,400 --> 00:52:23,800
Speaker 3: your kids?

989
00:52:24,440 --> 00:52:27,560
Speaker 4: The pointless exercise TVD. We're still on the journey, and

990
00:52:27,600 --> 00:52:30,400
Speaker 4: I think we will be for a long time. I

991
00:52:30,400 --> 00:52:33,319
Speaker 4: don't know how are you already using AI in your

992
00:52:33,360 --> 00:52:39,200
Speaker 4: day to day life today? Personally, I would say it's

993
00:52:39,239 --> 00:52:42,040
Speaker 4: replacing a good chunk of my search. You know, I'm

994
00:52:42,080 --> 00:52:45,160
Speaker 4: less likely to go blindly stumbling through a bunch of

995
00:52:45,160 --> 00:52:48,200
Speaker 4: web pages looking for stuff. I'm more likely to ask

996
00:52:48,239 --> 00:52:51,240
Speaker 4: a question from a few AI engines kind of see

997
00:52:51,400 --> 00:52:53,360
Speaker 4: get me in the right direction. Then I'll go bumble

998
00:52:53,400 --> 00:52:56,520
Speaker 4: through a few things. At work, I can tell you

999
00:52:57,640 --> 00:53:02,560
Speaker 4: code development right now, we are seeing massive improvements in

1000
00:53:02,680 --> 00:53:06,239
Speaker 4: code development and support. Products we have like Watson Code

1001
00:53:06,280 --> 00:53:11,719
Speaker 4: Assistant that is really showing immediate return for a code developers,

1002
00:53:11,719 --> 00:53:14,919
Speaker 4: and I think that will again be a tool that

1003
00:53:15,040 --> 00:53:20,120
Speaker 4: increases productivity for code developers immediately across the globe. Yeah.

1004
00:53:20,239 --> 00:53:23,600
Speaker 3: Last question, what's the one skill that every technology leader

1005
00:53:23,760 --> 00:53:26,000
Speaker 3: needs that has nothing to do with technology.

1006
00:53:26,920 --> 00:53:31,040
Speaker 4: Being able to inspire a set of people toward a

1007
00:53:31,080 --> 00:53:34,600
Speaker 4: common goal and collaborate to achieve it. That's at the

1008
00:53:34,640 --> 00:53:38,440
Speaker 4: core of everything everything. That's a lovely way to end.

1009
00:53:39,080 --> 00:53:40,440
Speaker 4: Thank you so much, Rick, Thank you.

1010
00:53:42,920 --> 00:53:47,240
Speaker 3: This conversation left me excited. I'm now imagining the potential

1011
00:53:47,280 --> 00:53:50,080
Speaker 3: for new use cases for AI in all sorts of

1012
00:53:50,120 --> 00:53:53,840
Speaker 3: different businesses. Rick didn't seem soldom my idea of a

1013
00:53:53,840 --> 00:53:57,279
Speaker 3: coffee shop makeover, but it's clear there's lots of opportunities

1014
00:53:57,360 --> 00:54:01,279
Speaker 3: here to increase speed and efficiency, to achieve your objectives,

1015
00:54:01,440 --> 00:54:04,960
Speaker 3: and to dream beyond the current applications for this technology.

1016
00:54:05,920 --> 00:54:08,480
Speaker 3: At the end of the day, the scaling of AI

1017
00:54:08,560 --> 00:54:11,920
Speaker 3: will rely on the right infrastructure to support it. With

1018
00:54:12,000 --> 00:54:15,120
Speaker 3: the right tools, you can solve problems that are unique

1019
00:54:15,160 --> 00:54:30,000
Speaker 3: tier industry and improve the experience for your customers. Smart

1020
00:54:30,000 --> 00:54:33,840
Speaker 3: Talks with IBM is produced by Matt Romano, Amy Gains McQuaid,

1021
00:54:34,080 --> 00:54:38,560
Speaker 3: and Jacob Goldstein. Were edited by Lydia gene Kott, mastering

1022
00:54:38,600 --> 00:54:43,759
Speaker 3: by Jake Koorsky. Theme song by Gramoscope. Special thanks to

1023
00:54:43,800 --> 00:54:46,319
Speaker 3: the eight Bar and IBM teams, as well as the

1024
00:54:46,400 --> 00:54:50,520
Speaker 3: Pushkin marketing team. Smart Talks with IBM is a production

1025
00:54:50,600 --> 00:54:56,799
Speaker 3: of Pushkin Industries and Ruby Studio at iHeartMedia. To find

1026
00:54:56,840 --> 00:55:01,960
Speaker 3: more Pushkin podcasts, listen on the iHeartRadio app, Apple Podcasts,

1027
00:55:02,080 --> 00:55:11,399
Speaker 3: or wherever you listen to podcasts. I'm Malcolm Gladwell. This

1028
00:55:11,600 --> 00:55:15,160
Speaker 3: is a paid advertisement from IBM. The conversations on this

1029
00:55:15,200 --> 00:55:32,640
Speaker 3: podcast don't necessarily represent IBM's positions, strategies, or opinions,