1
00:00:02,960 --> 00:00:09,680
Speaker 1: Bloomberg Audio Studios, podcasts, radio news. Now, experts have struggle

2
00:00:09,720 --> 00:00:11,760
Speaker 1: to get an accurate picture of just how many jobs

3
00:00:11,800 --> 00:00:15,920
Speaker 1: will be eliminated as artificial intelligence advances. One tracker has

4
00:00:15,960 --> 00:00:18,760
Speaker 1: actually reported US firms have announced to four thousand, six

5
00:00:18,880 --> 00:00:22,079
Speaker 1: hundred job cuts since may relate it to artificial intelligence,

6
00:00:22,280 --> 00:00:25,440
Speaker 1: but there are fears that this could be a vast underestimate. Now,

7
00:00:25,440 --> 00:00:27,840
Speaker 1: of course, one of the companies at the forefront, very

8
00:00:27,920 --> 00:00:30,880
Speaker 1: forefront of AI development at Google, and the firm just

9
00:00:30,880 --> 00:00:33,560
Speaker 1: announced twenty five million euros of funding to support AI

10
00:00:33,600 --> 00:00:36,320
Speaker 1: training and skills for people across at Europe. Well, I'm

11
00:00:36,400 --> 00:00:38,720
Speaker 1: very pleased to be joined for an exclusive conversation by

12
00:00:38,800 --> 00:00:41,320
Speaker 1: Google's president of Business and Operations, and you're the Middle

13
00:00:41,320 --> 00:00:43,519
Speaker 1: East and Africa, not Britain. Matt As always, thank you

14
00:00:43,600 --> 00:00:46,199
Speaker 1: so much for joining us. When you look at some

15
00:00:46,280 --> 00:00:48,800
Speaker 1: of the changes in AI and what that means for

16
00:00:48,920 --> 00:00:52,440
Speaker 1: job losses, can you quantify it and what does retraining

17
00:00:52,479 --> 00:00:53,199
Speaker 1: actually look like?

18
00:00:54,400 --> 00:00:57,040
Speaker 2: Yeah, I mean I think there's an array of analysis

19
00:00:57,040 --> 00:00:59,000
Speaker 2: out there, isn't there, But the most important thing is

20
00:00:59,040 --> 00:01:03,400
Speaker 2: that we make sure that is left behind behind by AI. Actually,

21
00:01:03,880 --> 00:01:07,280
Speaker 2: an estimate of one point two trillion euros one point

22
00:01:07,319 --> 00:01:09,600
Speaker 2: to trillion euros of economic growth in Europe if we

23
00:01:09,680 --> 00:01:12,280
Speaker 2: land it. Well, it's definitely the case that the work

24
00:01:12,319 --> 00:01:15,120
Speaker 2: we do will change. For most people, what they do

25
00:01:15,160 --> 00:01:18,160
Speaker 2: within their jobs is likely to shift. For some people

26
00:01:18,480 --> 00:01:20,840
Speaker 2: maybe the work will go away, and those are the

27
00:01:20,840 --> 00:01:22,760
Speaker 2: ones we really need to make sure we retrain. But

28
00:01:22,760 --> 00:01:25,160
Speaker 2: then there'll be whole new industries. You know, when I

29
00:01:25,240 --> 00:01:27,880
Speaker 2: left university, the web haven't been invented, and for the

30
00:01:27,959 --> 00:01:30,200
Speaker 2: last twenty years I've worked in the job that depended

31
00:01:30,240 --> 00:01:33,360
Speaker 2: on it. We didn't have web designers and data analysts

32
00:01:33,400 --> 00:01:35,240
Speaker 2: and so on. Then, so I think there's an overall

33
00:01:35,360 --> 00:01:39,959
Speaker 2: huge opportunity. And the reason we're launching this Opportunity initiative

34
00:01:40,120 --> 00:01:42,520
Speaker 2: for Europe with twenty five million euros is to make

35
00:01:42,560 --> 00:01:44,520
Speaker 2: sure that we don't leave anybody behind.

36
00:01:45,520 --> 00:01:47,840
Speaker 1: Yeah, but Matt, I guess the difficulty is you don't

37
00:01:47,880 --> 00:01:50,280
Speaker 1: really know where we'll end up, so we don't really

38
00:01:50,320 --> 00:01:52,960
Speaker 1: know the skills needed for tomorrow. Do you do this

39
00:01:53,040 --> 00:01:56,320
Speaker 1: with governments? How do you see? If I tell you, look, Matt, Britain,

40
00:01:56,440 --> 00:01:58,760
Speaker 1: what does the workforce look like in five and then

41
00:01:58,800 --> 00:02:02,360
Speaker 1: ten years? How do you answer that? Yeah?

42
00:02:02,440 --> 00:02:04,280
Speaker 2: Well, the good news is We've had some experience here

43
00:02:04,320 --> 00:02:06,800
Speaker 2: so and I took this role eight years ago there

44
00:02:06,840 --> 00:02:09,040
Speaker 2: was a digital skills gap in Europe and we set

45
00:02:09,080 --> 00:02:12,079
Speaker 2: out to try and train a million Europeans in digital skills.

46
00:02:12,400 --> 00:02:15,280
Speaker 2: Now eight years later, we've trained twelve million, and we've

47
00:02:15,280 --> 00:02:19,440
Speaker 2: worked with government ministries of labor, small business associations, trade

48
00:02:19,560 --> 00:02:22,400
Speaker 2: unions and others to bring digital skills to everyone. And

49
00:02:22,480 --> 00:02:25,440
Speaker 2: that's why we're confident in this twenty five million euro

50
00:02:25,639 --> 00:02:28,519
Speaker 2: Opportunity initiative that we're going to try to reach out

51
00:02:28,520 --> 00:02:30,880
Speaker 2: to those people who are perhaps the most vulnerable or

52
00:02:30,880 --> 00:02:34,680
Speaker 2: in underserved communities. Ten millions euro straightaway is going towards

53
00:02:34,720 --> 00:02:38,000
Speaker 2: reaching workers who are likely to be most vulnerable. We're

54
00:02:38,000 --> 00:02:42,560
Speaker 2: also boosting our growth academy for startups, which is most

55
00:02:42,560 --> 00:02:45,480
Speaker 2: people affected by AI will be an AI using economy.

56
00:02:45,600 --> 00:02:48,200
Speaker 2: Startups are likely to be the AI building economy, So

57
00:02:48,240 --> 00:02:51,000
Speaker 2: we're launching an effort with health to begin with there

58
00:02:51,360 --> 00:02:54,400
Speaker 2: and then building on our digital skills stuff. We've got

59
00:02:54,480 --> 00:02:57,960
Speaker 2: AI fundamentals for everyone and we're making that available in

60
00:02:58,040 --> 00:03:00,520
Speaker 2: eighteen more languages today. So I think the right thing

61
00:03:00,560 --> 00:03:02,560
Speaker 2: to do here is say that AI has a huge

62
00:03:02,560 --> 00:03:06,040
Speaker 2: opportunity to help all of us in science, health, work,

63
00:03:06,160 --> 00:03:08,600
Speaker 2: education and life. But there's a risk that some are

64
00:03:08,639 --> 00:03:10,560
Speaker 2: left behind, and that's why we need to lean in

65
00:03:10,639 --> 00:03:13,320
Speaker 2: together with governments and others to make sure that we're

66
00:03:13,320 --> 00:03:15,440
Speaker 2: equip for the future. You're right, we don't know exactly

67
00:03:15,480 --> 00:03:17,520
Speaker 2: what those jobs are, but we do know people will

68
00:03:17,800 --> 00:03:19,799
Speaker 2: need to be able to be get confident in using

69
00:03:19,880 --> 00:03:20,360
Speaker 2: the schools.

70
00:03:22,120 --> 00:03:24,959
Speaker 1: How does an AI regulation in Europe actually affect I

71
00:03:24,960 --> 00:03:27,919
Speaker 1: guess you're a development and roll out of AI products

72
00:03:28,400 --> 00:03:29,600
Speaker 1: versus the rest of the world.

73
00:03:30,560 --> 00:03:33,200
Speaker 2: Yeah. I think what we see in Europe is incredible

74
00:03:33,200 --> 00:03:37,040
Speaker 2: skills and desire for AI. When you survey people, the

75
00:03:37,160 --> 00:03:40,440
Speaker 2: vast majority of Europeans believe that AI can have a

76
00:03:41,080 --> 00:03:44,720
Speaker 2: benefit to them and to society, so there's a desire

77
00:03:44,720 --> 00:03:47,080
Speaker 2: to see it landed. But what the surveys tell us

78
00:03:47,080 --> 00:03:50,280
Speaker 2: across seventeen thousand Europeans and others is that they want

79
00:03:50,320 --> 00:03:53,120
Speaker 2: governments and technology companies to work together to make the

80
00:03:53,160 --> 00:03:57,520
Speaker 2: technology safe and accessible to everyone. Now, today's initiative is

81
00:03:57,560 --> 00:04:00,120
Speaker 2: making sure that it's accessible to everyone.

82
00:04:00,160 --> 00:04:00,520
Speaker 1: It's safe.

83
00:04:00,520 --> 00:04:03,440
Speaker 2: That's for the regulators to set out the guardrails and

84
00:04:03,480 --> 00:04:06,120
Speaker 2: the AI Act in Europe. The ink is just drawing

85
00:04:06,200 --> 00:04:08,240
Speaker 2: on that, and that has been a two or three

86
00:04:08,280 --> 00:04:11,200
Speaker 2: year engagement with companies like ours and communities to try

87
00:04:11,200 --> 00:04:13,440
Speaker 2: to come up with good rules of the road. And

88
00:04:13,440 --> 00:04:15,560
Speaker 2: now the devil's in the detail of how we apply that.

89
00:04:15,600 --> 00:04:18,280
Speaker 2: But we think it's such an important technology it needs

90
00:04:18,279 --> 00:04:20,599
Speaker 2: to be regulated so that we can harness it for

91
00:04:20,680 --> 00:04:21,480
Speaker 2: good for everyone.

92
00:04:23,400 --> 00:04:25,000
Speaker 1: And is it really if you look at, for example,

93
00:04:25,000 --> 00:04:28,200
Speaker 1: your cloud business, is it all about AI that's driving

94
00:04:28,279 --> 00:04:30,120
Speaker 1: demand or is it a little bit more balanced.

95
00:04:31,360 --> 00:04:32,800
Speaker 2: Yeah, there's an array of things, but I think a

96
00:04:32,800 --> 00:04:35,320
Speaker 2: couple of things. I'd say that. You know, it's easy

97
00:04:35,360 --> 00:04:38,040
Speaker 2: for people to think that AI and chatbots are the

98
00:04:38,080 --> 00:04:40,920
Speaker 2: same thing. Actually AI has been around for much longer,

99
00:04:41,560 --> 00:04:44,400
Speaker 2: and it's much more than chatbots. So you know, seven

100
00:04:44,480 --> 00:04:46,599
Speaker 2: or eight years ago at Google we'd tried to pivot

101
00:04:46,640 --> 00:04:50,240
Speaker 2: to be an AI first company, and Google Translates is

102
00:04:50,240 --> 00:04:52,600
Speaker 2: perhaps where it was born. You know. The connection of

103
00:04:52,720 --> 00:04:56,680
Speaker 2: languages and understanding languages led to technical breakthroughs and these

104
00:04:56,680 --> 00:04:59,360
Speaker 2: things called large language models, which are now powering lots

105
00:04:59,360 --> 00:05:02,440
Speaker 2: of the generative AIU see. But AI is useful for

106
00:05:02,560 --> 00:05:04,960
Speaker 2: so much and the most exciting areas I think for

107
00:05:05,080 --> 00:05:08,920
Speaker 2: society are scientific and health breakthrough. So if I take

108
00:05:09,560 --> 00:05:12,120
Speaker 2: researchers into vaccines that's been top of mind for some time,

109
00:05:12,240 --> 00:05:16,520
Speaker 2: or drugs or crop resilience, that depends on understanding proteins.

110
00:05:17,000 --> 00:05:18,880
Speaker 2: And a few years ago, there are about one hundred

111
00:05:18,880 --> 00:05:22,960
Speaker 2: and seventy five thousand proteins that have been painstakingly identified

112
00:05:22,960 --> 00:05:25,760
Speaker 2: in three D structure by PhD students. It would take

113
00:05:25,800 --> 00:05:29,600
Speaker 2: five years for one PhD student to find one three

114
00:05:29,640 --> 00:05:32,840
Speaker 2: D structure of a protein. Now alpha fold, built by

115
00:05:32,920 --> 00:05:35,480
Speaker 2: my colleagues at Google d Mind, changed all that and

116
00:05:35,520 --> 00:05:38,960
Speaker 2: in a matter of months they catalog two hundred million proteins.

117
00:05:39,240 --> 00:05:43,240
Speaker 2: They're available for free to every expert working in this area,

118
00:05:43,320 --> 00:05:46,840
Speaker 2: and now one point seven million experts are using alpha

119
00:05:46,839 --> 00:05:51,240
Speaker 2: fold and that database of proteins to advance drug discovery,

120
00:05:51,279 --> 00:05:54,040
Speaker 2: disease research, crop resilience research. So I think things like

121
00:05:54,080 --> 00:05:56,719
Speaker 2: that which maybe people don't see when they see the

122
00:05:56,720 --> 00:05:59,640
Speaker 2: headlines about AI, that are changing what's going to be

123
00:05:59,640 --> 00:06:00,840
Speaker 2: possible about methods