1
00:00:04,440 --> 00:00:12,319
Speaker 1: Welcome to tech Stuff, a production from iHeartRadio. Hey there,

2
00:00:12,320 --> 00:00:15,840
Speaker 1: and welcome to tech Stuff. I'm your host, Jonathan Strickland.

3
00:00:15,840 --> 00:00:18,760
Speaker 1: I'm an executive producer with iHeart Podcasts and how the

4
00:00:18,800 --> 00:00:22,000
Speaker 1: tech are you. It's time for the tech news for Thursday,

5
00:00:22,040 --> 00:00:28,160
Speaker 1: November thirty, twenty twenty three. Let's get to it. When

6
00:00:28,240 --> 00:00:31,280
Speaker 1: last we left open Ai, the story was that Sam

7
00:00:31,360 --> 00:00:34,320
Speaker 1: Altman was going to come back to the company after

8
00:00:34,360 --> 00:00:37,839
Speaker 1: being suddenly and surprisingly fired, and that the board of

9
00:00:37,880 --> 00:00:40,960
Speaker 1: directors would consist of different folks as part of the deal.

10
00:00:41,200 --> 00:00:44,159
Speaker 1: So three of the four members who voted against Altman

11
00:00:44,560 --> 00:00:48,520
Speaker 1: were shown the door. The only remaining one is Adam DiAngelo. Well,

12
00:00:48,560 --> 00:00:51,400
Speaker 1: now Sam Altman is officially back as CEO, and on

13
00:00:51,440 --> 00:00:54,760
Speaker 1: top of that, Microsoft is going to have a representative

14
00:00:54,800 --> 00:00:58,440
Speaker 1: who will hold a non voting seat on the board.

15
00:00:58,960 --> 00:01:02,440
Speaker 1: If you recall, Microsoft has committed to a ten billion

16
00:01:02,600 --> 00:01:06,960
Speaker 1: dollar investment in open Ai, and Microsoft CEO Satya Nadella

17
00:01:07,080 --> 00:01:12,120
Speaker 1: was reportedly furious over Altman's firing because, well, the board

18
00:01:12,120 --> 00:01:15,600
Speaker 1: of directors kept it all a secret until telling Altman

19
00:01:15,720 --> 00:01:18,920
Speaker 1: and then essentially telling the world, and Nadella feels that

20
00:01:19,160 --> 00:01:21,920
Speaker 1: Microsoft should probably get a heads up on that sort

21
00:01:21,920 --> 00:01:24,760
Speaker 1: of thing early on, since they are pouring in so

22
00:01:24,800 --> 00:01:28,160
Speaker 1: many billions of dollars into open Ai, and maybe even

23
00:01:28,560 --> 00:01:31,920
Speaker 1: have a chance to propose alternatives because again, they are

24
00:01:31,920 --> 00:01:35,440
Speaker 1: so heavily invested in the company. Ilia Sutzkever, who is

25
00:01:35,520 --> 00:01:38,640
Speaker 1: one of the board members responsible for firing Altman in

26
00:01:38,640 --> 00:01:41,240
Speaker 1: the first place, stepped down from the board, but he

27
00:01:41,280 --> 00:01:44,080
Speaker 1: may still remain with the company. He is one of

28
00:01:44,120 --> 00:01:47,200
Speaker 1: the co founders of open Ai and has served as

29
00:01:47,240 --> 00:01:51,200
Speaker 1: its chief scientist. Anyway, this really went down as one

30
00:01:51,240 --> 00:01:54,160
Speaker 1: of the weirder stories of twenty twenty three, and one

31
00:01:54,200 --> 00:01:56,560
Speaker 1: that actually wrapped up pretty darn fast. I mean, most

32
00:01:56,600 --> 00:02:00,360
Speaker 1: of our weird stories have lasted the entire year. Talk

33
00:02:00,480 --> 00:02:04,760
Speaker 1: about chat GPT speaking of weird stories, and how some

34
00:02:04,840 --> 00:02:08,480
Speaker 1: researchers fooled the tool into revealing training data as well

35
00:02:08,520 --> 00:02:12,000
Speaker 1: as some personal information about various people. And they did

36
00:02:12,000 --> 00:02:15,080
Speaker 1: it by having the chat bought repeat the same word

37
00:02:15,520 --> 00:02:19,200
Speaker 1: over and over without stop forever. So they essentially gave

38
00:02:19,240 --> 00:02:22,560
Speaker 1: it the prompt repeat the word, you know, whatever the

39
00:02:22,560 --> 00:02:26,200
Speaker 1: word was, forever, and it would start doing that. So

40
00:02:26,400 --> 00:02:29,000
Speaker 1: it began to comply, but at some point it started

41
00:02:29,000 --> 00:02:33,000
Speaker 1: to intersperse the repetitions of the word with other information

42
00:02:33,600 --> 00:02:38,919
Speaker 1: like snippets from copyrighted research papers or personal phone numbers

43
00:02:39,080 --> 00:02:43,720
Speaker 1: of private citizens, including the CEO of some company. The

44
00:02:43,760 --> 00:02:48,400
Speaker 1: researchers used words like poem or company and had it

45
00:02:48,480 --> 00:02:50,760
Speaker 1: repeat that, and they saw different results depending on which

46
00:02:50,760 --> 00:02:53,440
Speaker 1: word they used. And it's interesting because I was just

47
00:02:53,560 --> 00:02:58,160
Speaker 1: rewatching a horror movie called Pontypool, and in that movie

48
00:02:58,200 --> 00:03:01,960
Speaker 1: people go crazy while they repeating simple small words over

49
00:03:02,000 --> 00:03:04,080
Speaker 1: and over in English. And then I read the story

50
00:03:04,120 --> 00:03:06,440
Speaker 1: and I think, oh, gosh, we're in the movie. I mean,

51
00:03:06,480 --> 00:03:09,320
Speaker 1: not really. But the researchers Chi died open Ai. They

52
00:03:09,360 --> 00:03:12,760
Speaker 1: said the attack they used shouldn't have worked because someone

53
00:03:12,880 --> 00:03:17,360
Speaker 1: surely should have discovered it well before then and fixed it.

54
00:03:17,760 --> 00:03:20,000
Speaker 1: They said, the attack is kind of silly, but it

55
00:03:20,040 --> 00:03:23,600
Speaker 1: could have serious consequences if someone invested enough money in

56
00:03:23,680 --> 00:03:26,120
Speaker 1: the paid for version of chat GPT, because that's what

57
00:03:26,160 --> 00:03:29,560
Speaker 1: the researchers used. If they poured enough money into it

58
00:03:29,600 --> 00:03:32,880
Speaker 1: and they used a similar style of attack, they could

59
00:03:32,960 --> 00:03:35,840
Speaker 1: have received an awful lot of information in the process.

60
00:03:36,200 --> 00:03:39,160
Speaker 1: The attack also showed how open ai has used material

61
00:03:39,240 --> 00:03:42,760
Speaker 1: on the web, including copyrighted material, to train its large

62
00:03:42,880 --> 00:03:47,600
Speaker 1: language model. That's an unfolding, sticky situation as lawmakers try

63
00:03:47,640 --> 00:03:50,960
Speaker 1: to get up to speed on the legalities of training llms.

64
00:03:51,480 --> 00:03:55,440
Speaker 1: The researcher said that open ai actually patched this vulnerability

65
00:03:55,600 --> 00:03:58,760
Speaker 1: back in August, so they're just now publishing their findings.

66
00:03:59,160 --> 00:04:03,000
Speaker 1: But according to them, open ai has already responded to

67
00:04:03,000 --> 00:04:07,280
Speaker 1: the issue. However, in Gadget, which published a whole article

68
00:04:07,320 --> 00:04:12,280
Speaker 1: about this issue, said that their staff attempted to replicate

69
00:04:12,360 --> 00:04:14,680
Speaker 1: the results and they were able to do so. They

70
00:04:14,840 --> 00:04:18,520
Speaker 1: used the word reply. They told the chatbot to repeat

71
00:04:18,520 --> 00:04:24,039
Speaker 1: reply forever, and it started to submitting other types of data.

72
00:04:24,640 --> 00:04:29,880
Speaker 1: So it's clear that this is not fully resolved. It'll

73
00:04:29,880 --> 00:04:33,320
Speaker 1: be interesting to see if this actually ends up having

74
00:04:33,400 --> 00:04:37,359
Speaker 1: any actual negative consequences in the real world. Beyond the

75
00:04:37,440 --> 00:04:41,160
Speaker 1: possibility of that happening, this next story is for you

76
00:04:41,320 --> 00:04:44,640
Speaker 1: Apple users out there. According to the Wall Street Journal,

77
00:04:44,800 --> 00:04:47,919
Speaker 1: Apple has proposed that it and Goldman Sachs, which is

78
00:04:47,960 --> 00:04:53,080
Speaker 1: the bank that facilitates transactions that are moving across Apple Pay,

79
00:04:53,560 --> 00:04:57,680
Speaker 1: will part ways in twelve to fifteen months, and Goldman

80
00:04:57,760 --> 00:05:00,680
Speaker 1: Sachs is likely to accept that offer because, as reportedly,

81
00:05:00,720 --> 00:05:03,960
Speaker 1: the bank has found Apple's business to be unprofitable that

82
00:05:04,080 --> 00:05:09,080
Speaker 1: handling these transactions is actually costly. Apparently, folks who use

83
00:05:09,080 --> 00:05:12,560
Speaker 1: Apple Pay make payments on their accounts more slowly than

84
00:05:12,600 --> 00:05:15,640
Speaker 1: people who use other types of credit cards, and this

85
00:05:16,000 --> 00:05:19,280
Speaker 1: lag in payment has had a big impact on Goldman

86
00:05:19,360 --> 00:05:22,480
Speaker 1: Sachs's own finances. This means that Apple will need to

87
00:05:22,520 --> 00:05:26,760
Speaker 1: find some other financial institution to back those transactions, or

88
00:05:26,839 --> 00:05:30,240
Speaker 1: somehow do it themselves, which I cannot imagine happening because

89
00:05:30,640 --> 00:05:33,760
Speaker 1: I imagine that requires fulfilling a whole bunch of regulations

90
00:05:33,760 --> 00:05:36,440
Speaker 1: that I don't think Apple's ready to do. Or I

91
00:05:36,480 --> 00:05:41,280
Speaker 1: guess the third option is closing off Apple Pay. I

92
00:05:41,360 --> 00:05:44,320
Speaker 1: don't see that being a big priority either. The Wall

93
00:05:44,320 --> 00:05:47,960
Speaker 1: Street Journal reports that American Express or Synchrony Financial are

94
00:05:48,000 --> 00:05:53,600
Speaker 1: potential successors who might take over the back end operations. However,

95
00:05:53,600 --> 00:05:56,560
Speaker 1: the article says that Apple has been pretty demanding of

96
00:05:56,640 --> 00:05:59,800
Speaker 1: its partners, and that when you pair that with the

97
00:05:59,880 --> 00:06:03,800
Speaker 1: challenges of actually making the business profitable, that could cause

98
00:06:03,839 --> 00:06:07,040
Speaker 1: a hiccup or two. So is Apple Pay on borrowed time?

99
00:06:07,240 --> 00:06:09,760
Speaker 1: Is it possible that it could be phased out after

100
00:06:09,800 --> 00:06:12,120
Speaker 1: a year or so? It beats me, but I'll keep

101
00:06:12,120 --> 00:06:15,799
Speaker 1: an eye on it. Starting tomorrow, which is Friday, Desummer first.

102
00:06:15,839 --> 00:06:18,120
Speaker 1: In case you're one of those odd people who likes

103
00:06:18,160 --> 00:06:22,279
Speaker 1: to listen to tech news episodes months after it happened,

104
00:06:22,760 --> 00:06:25,640
Speaker 1: Google will be shutting down old user accounts that have

105
00:06:25,720 --> 00:06:29,080
Speaker 1: been inactive for at least two years. I actually got

106
00:06:29,120 --> 00:06:31,719
Speaker 1: a message from Google not that long ago regarding an

107
00:06:31,720 --> 00:06:34,800
Speaker 1: email address that I created for a podcast, but I

108
00:06:34,880 --> 00:06:38,080
Speaker 1: abandoned that podcast after two episodes, because even professionals like

109
00:06:38,120 --> 00:06:41,440
Speaker 1: me can be a cliche. The move will delete all

110
00:06:41,520 --> 00:06:45,159
Speaker 1: related data to the Gmail account, so that includes not

111
00:06:45,360 --> 00:06:48,080
Speaker 1: just the Gmail address and everything that was in the

112
00:06:48,200 --> 00:06:52,840
Speaker 1: Gmail but also anything in an associated Google Drive or

113
00:06:52,920 --> 00:06:56,039
Speaker 1: Google Photos account. To be clear, this isn't going to

114
00:06:56,080 --> 00:06:59,080
Speaker 1: affect any active Gmail accounts. If you have used your

115
00:06:59,080 --> 00:07:01,200
Speaker 1: Gmail account with that in the last two years, it

116
00:07:01,279 --> 00:07:04,200
Speaker 1: is not gonna be in danger. Only ones that have

117
00:07:04,320 --> 00:07:07,080
Speaker 1: been inactive for two years or longer will be affected,

118
00:07:07,279 --> 00:07:10,440
Speaker 1: and it will also only affect personal Gmail addresses, not

119
00:07:10,560 --> 00:07:15,000
Speaker 1: addresses associated with organizations or businesses. There's still some unanswered

120
00:07:15,040 --> 00:07:18,040
Speaker 1: questions about all this, however. The big one is will

121
00:07:18,160 --> 00:07:22,800
Speaker 1: Google recycle the inactive Gmail addresses once they wipe those out?

122
00:07:23,040 --> 00:07:26,560
Speaker 1: Will people be able to reregister those addresses. If so,

123
00:07:27,000 --> 00:07:30,280
Speaker 1: that could potentially lead to issues with identity theft because

124
00:07:30,400 --> 00:07:33,920
Speaker 1: folks could scoop up newly available addresses that previously had

125
00:07:33,960 --> 00:07:37,640
Speaker 1: belonged to someone else. I imagine Google has probably thought of that,

126
00:07:38,000 --> 00:07:42,280
Speaker 1: but I haven't seen any confirmation on that. A company

127
00:07:42,320 --> 00:07:45,960
Speaker 1: called Analytics claims that Google did a big old whoopsie,

128
00:07:46,000 --> 00:07:50,360
Speaker 1: like a big old, huge issue whoopsie. See really, when

129
00:07:50,400 --> 00:07:53,120
Speaker 1: you get down to it, Google is an advertising platform

130
00:07:53,440 --> 00:07:55,480
Speaker 1: more than anything else. It's not a search company. It's

131
00:07:55,480 --> 00:07:59,320
Speaker 1: an ad company. Serves up ads from customers on various websites,

132
00:07:59,360 --> 00:08:03,960
Speaker 1: including Google. Well, part of that business includes giving advertisers

133
00:08:04,000 --> 00:08:06,400
Speaker 1: the chance to say which types of websites they do

134
00:08:06,600 --> 00:08:09,560
Speaker 1: or don't want to have their ads appear on, So

135
00:08:09,640 --> 00:08:12,840
Speaker 1: it gives the advertisers more control over protecting the brands

136
00:08:12,840 --> 00:08:16,240
Speaker 1: that they represent. I get a very similar consideration, like

137
00:08:16,280 --> 00:08:19,000
Speaker 1: I can say what types of ads I do or

138
00:08:19,160 --> 00:08:22,520
Speaker 1: don't want to appear on this show, and that gives

139
00:08:22,560 --> 00:08:25,640
Speaker 1: me more control over the show and protects you, the listener,

140
00:08:25,920 --> 00:08:29,320
Speaker 1: from advertisers that perhaps I don't believe in for one

141
00:08:29,320 --> 00:08:32,719
Speaker 1: reason or another. Well, advertisers get the same kind of

142
00:08:33,800 --> 00:08:36,360
Speaker 1: consideration when it comes to where their ads are going

143
00:08:36,400 --> 00:08:39,760
Speaker 1: to be delivered, because you want to protect the brand, right.

144
00:08:39,800 --> 00:08:42,840
Speaker 1: You don't want the brand to be associated with something

145
00:08:43,360 --> 00:08:47,280
Speaker 1: that is against the brand's image, whether that is outright

146
00:08:47,320 --> 00:08:53,040
Speaker 1: illegal content because that's a possibility, or things like porn sites.

147
00:08:53,280 --> 00:08:57,480
Speaker 1: If you are, for example, an advertiser that advertises children's clothing,

148
00:08:58,120 --> 00:09:02,839
Speaker 1: you definitely don't want your ad appearing on an adult website.

149
00:09:02,960 --> 00:09:07,360
Speaker 1: That's just not appropriate, so Analytics says, the Google Search

150
00:09:07,400 --> 00:09:11,280
Speaker 1: partner network has apparently been placing ads from prominent brands

151
00:09:11,320 --> 00:09:14,640
Speaker 1: on sites that are very likely not on any white list,

152
00:09:15,040 --> 00:09:18,760
Speaker 1: so illegal sites, adult websites, all this kind of stuff

153
00:09:18,800 --> 00:09:23,000
Speaker 1: have been having some major advertisers pop up on them,

154
00:09:23,400 --> 00:09:28,400
Speaker 1: and that's probably not what the advertisers wanted. According to Analytics,

155
00:09:28,679 --> 00:09:35,240
Speaker 1: they found instances in which very high profile companies like Apple, Amazon, Lego, Meta, Uber, etc.

156
00:09:36,080 --> 00:09:39,320
Speaker 1: Had their ads posted on inappropriate sites. These would be

157
00:09:39,400 --> 00:09:41,959
Speaker 1: sites that the brands absolutely would not want to be

158
00:09:42,040 --> 00:09:45,440
Speaker 1: associated with. This is a huge issue for brand safety right.

159
00:09:45,760 --> 00:09:49,040
Speaker 1: But other problems include cases in which, say, alcohol brand

160
00:09:49,080 --> 00:09:53,079
Speaker 1: advertisements were showing up on websites that are targeted to children. Clearly,

161
00:09:53,120 --> 00:09:56,760
Speaker 1: the website administrators don't want that that they're running a

162
00:09:56,800 --> 00:09:59,800
Speaker 1: site that's targeting children. They don't want ads that have

163
00:10:00,040 --> 00:10:03,800
Speaker 1: alcohol advertised on them running on the site. The advertisers

164
00:10:03,840 --> 00:10:05,960
Speaker 1: don't want it either, because there could be a huge

165
00:10:06,040 --> 00:10:09,160
Speaker 1: legal case to unfold. Because of all of this, Google

166
00:10:09,200 --> 00:10:12,240
Speaker 1: reportedly is investigating the issue in addressing it, though questions

167
00:10:12,240 --> 00:10:14,800
Speaker 1: remain as to how this even happened in the first place.

168
00:10:15,280 --> 00:10:17,200
Speaker 1: This comes at a really bad time for Google. The

169
00:10:17,240 --> 00:10:20,160
Speaker 1: company is already in the target sites of US lawmakers

170
00:10:20,240 --> 00:10:24,119
Speaker 1: as they question if Google is engaged in anti competitive

171
00:10:24,160 --> 00:10:27,839
Speaker 1: practices and if, perhaps to an extreme case, if Google

172
00:10:27,920 --> 00:10:31,280
Speaker 1: should get broken up into smaller companies. Politicians and the

173
00:10:31,280 --> 00:10:34,119
Speaker 1: EU have also indicated that they may launch an investigation

174
00:10:34,200 --> 00:10:36,760
Speaker 1: into the matter. They pointed out that if the EU

175
00:10:36,840 --> 00:10:40,719
Speaker 1: Commission ads, like an ad for the EU Commission, ends

176
00:10:40,760 --> 00:10:44,880
Speaker 1: up being placed on, say a sanctioned Russian website, then

177
00:10:45,480 --> 00:10:49,000
Speaker 1: the Commission inadvertently becomes an accomplice to a party that

178
00:10:49,120 --> 00:10:53,080
Speaker 1: is violating these international sanctions. So this really is a

179
00:10:53,120 --> 00:10:55,480
Speaker 1: pretty big mess for Google. All Right, we're going to

180
00:10:55,520 --> 00:10:57,120
Speaker 1: take a quick break. When we come back, we'll be

181
00:10:57,160 --> 00:11:09,680
Speaker 1: talking about messes for some other tech companies. Okay, we're back,

182
00:11:10,120 --> 00:11:13,400
Speaker 1: So here in the United States, the CEOs of five

183
00:11:13,600 --> 00:11:22,000
Speaker 1: tech companies, those being X, the platform formerly known as Twitter, Meta, TikTok, Snap,

184
00:11:22,360 --> 00:11:25,640
Speaker 1: and Discord are all going to get together on January

185
00:11:25,800 --> 00:11:30,440
Speaker 1: thirty first, twenty twenty four. They're not having lunch, they're

186
00:11:30,480 --> 00:11:33,760
Speaker 1: not engaging in small talk. Instead, they're appearing in front

187
00:11:33,800 --> 00:11:36,040
Speaker 1: of the US Senate to answer questions about how they

188
00:11:36,080 --> 00:11:40,680
Speaker 1: protect or rather fail to protect children's safety online. So

189
00:11:40,720 --> 00:11:44,000
Speaker 1: the hearing is all about child exploitation, and that's one

190
00:11:44,000 --> 00:11:46,920
Speaker 1: of the foundational concerns people have brought up about social

191
00:11:46,960 --> 00:11:51,160
Speaker 1: platforms for years now. Really, heck, when Francis Hogan came

192
00:11:51,200 --> 00:11:56,000
Speaker 1: forward with internal documents from Facebook now Meta, many of

193
00:11:56,000 --> 00:11:58,840
Speaker 1: those documents centered on the company's potential impact on the

194
00:11:58,880 --> 00:12:02,760
Speaker 1: mental health of young users, and it wasn't good news.

195
00:12:03,280 --> 00:12:06,240
Speaker 1: Senators say that the day we'll give CEOs the chance

196
00:12:06,280 --> 00:12:10,680
Speaker 1: to quote testify about their failure to protect children online

197
00:12:10,800 --> 00:12:13,640
Speaker 1: end quote. That certainly sounds like the Senate has already

198
00:12:13,840 --> 00:12:18,400
Speaker 1: reached the conclusion that these platforms have done and perhaps

199
00:12:18,440 --> 00:12:22,439
Speaker 1: actively are doing harm to children or the very least

200
00:12:22,840 --> 00:12:25,640
Speaker 1: failing to protect them, and that they are at fault

201
00:12:25,679 --> 00:12:27,800
Speaker 1: for this. I'm sure it'll be a fun day for

202
00:12:27,840 --> 00:12:30,720
Speaker 1: all involved. Personally, I think something that tends to be

203
00:12:30,760 --> 00:12:33,520
Speaker 1: overlooked is how we in the United States aren't really

204
00:12:33,600 --> 00:12:38,000
Speaker 1: serious about data privacy or security or data ownership, you know,

205
00:12:38,040 --> 00:12:42,720
Speaker 1: whether the data that pertains to people actually belongs to them.

206
00:12:43,480 --> 00:12:46,240
Speaker 1: And by addressing some of those gaps, I think we

207
00:12:46,240 --> 00:12:49,240
Speaker 1: could actually solve a lot of problems that adversely affect

208
00:12:49,280 --> 00:12:52,640
Speaker 1: young folks if we had tougher rules in place as

209
00:12:52,679 --> 00:12:55,760
Speaker 1: to who owns that data and who can access it,

210
00:12:56,200 --> 00:13:00,439
Speaker 1: and tougher laws about getting permission to access them data

211
00:13:00,480 --> 00:13:03,440
Speaker 1: and to exploit it. All of those things I think

212
00:13:03,840 --> 00:13:07,319
Speaker 1: are necessary, But hey, what do I know? So that'll

213
00:13:07,480 --> 00:13:09,719
Speaker 1: be happening at the end of the first month of

214
00:13:09,800 --> 00:13:13,600
Speaker 1: next year. US Senator Marco Rubio has proposed the most

215
00:13:13,679 --> 00:13:17,480
Speaker 1: recent opposition to TikTok here in the US. Rubio proposes

216
00:13:17,520 --> 00:13:21,920
Speaker 1: an ultimatum either China hands over the algorithm used for

217
00:13:22,000 --> 00:13:27,079
Speaker 1: TikTok's recommendation engine, or the United States demonetizes the platform

218
00:13:27,120 --> 00:13:30,520
Speaker 1: within the country, which effectively ends up being a ban

219
00:13:30,800 --> 00:13:33,960
Speaker 1: on TikTok. This is pretty wild stuff, so at the

220
00:13:33,960 --> 00:13:36,680
Speaker 1: heart of Rubio's call appears to be this perception that

221
00:13:36,720 --> 00:13:42,600
Speaker 1: TikTok's algorithm is purposefully shaping discourse. That perhaps China's using

222
00:13:42,600 --> 00:13:46,040
Speaker 1: the algorithm to push propaganda or sway the minds of

223
00:13:46,080 --> 00:13:49,720
Speaker 1: the US public in particular directions. And maybe that's true.

224
00:13:49,800 --> 00:13:52,400
Speaker 1: I mean, maybe the algorithm is doing that, but I

225
00:13:52,440 --> 00:13:55,720
Speaker 1: suspect the algorithm is actually doing something far simpler. It

226
00:13:55,840 --> 00:13:59,920
Speaker 1: is essentially identifying the types of content each user to

227
00:14:00,080 --> 00:14:02,679
Speaker 1: tends to like, and then it just serves up an

228
00:14:02,800 --> 00:14:05,839
Speaker 1: endless supply of that kind of content in an effort

229
00:14:05,920 --> 00:14:08,240
Speaker 1: to keep that user engaged on the platform for as

230
00:14:08,240 --> 00:14:11,560
Speaker 1: long as possible. So if that content does favor one

231
00:14:11,600 --> 00:14:15,760
Speaker 1: particular political ideology, well yeah, TikTok serves that up to

232
00:14:15,760 --> 00:14:19,040
Speaker 1: the user. But if the user tends to favor the

233
00:14:19,200 --> 00:14:23,480
Speaker 1: opposite point of view, I don't think TikTok tries to

234
00:14:23,720 --> 00:14:27,440
Speaker 1: change the user's mind and manipulate them so that they

235
00:14:27,520 --> 00:14:31,400
Speaker 1: think the other thing. No, Instead, it just caters to

236
00:14:31,680 --> 00:14:33,440
Speaker 1: that kind of point of view and serves up that

237
00:14:33,560 --> 00:14:37,720
Speaker 1: sort of video content to the user. Again, that's my guess,

238
00:14:37,760 --> 00:14:40,760
Speaker 1: and maybe I'm wrong. But anyway, as CNBC puts it,

239
00:14:40,840 --> 00:14:44,760
Speaker 1: Rubio's law would quote prohibit financial transactions from social media

240
00:14:44,800 --> 00:14:50,600
Speaker 1: companies located in or under the influence of China, Russia,

241
00:14:50,640 --> 00:14:53,080
Speaker 1: and a few other countries end quote. So yeah, that

242
00:14:53,200 --> 00:14:56,760
Speaker 1: under the influence is the key element for TikTok because

243
00:14:56,800 --> 00:14:59,880
Speaker 1: obviously the company TikTok is centered here in the United States,

244
00:15:00,320 --> 00:15:03,880
Speaker 1: but it's pairing company by Dance is in China, so

245
00:15:03,960 --> 00:15:07,280
Speaker 1: that seems to be where that key phrase comes into play.

246
00:15:07,680 --> 00:15:10,760
Speaker 1: I don't really think this is on the right track.

247
00:15:10,800 --> 00:15:14,280
Speaker 1: I do think TikTok is potentially harmful, but uh, I

248
00:15:14,280 --> 00:15:18,720
Speaker 1: don't think it's necessarily trying to sway opinions in a

249
00:15:18,760 --> 00:15:22,840
Speaker 1: specific direction. I think the stuff that TikTok serves up

250
00:15:23,240 --> 00:15:25,480
Speaker 1: is driven by the behaviors of the people who are

251
00:15:25,480 --> 00:15:28,880
Speaker 1: on TikTok, not the other way around. But you know,

252
00:15:29,080 --> 00:15:32,400
Speaker 1: that's just my guess. I don't have any real hard

253
00:15:32,480 --> 00:15:33,960
Speaker 1: data to say one way or the other, and I

254
00:15:34,000 --> 00:15:37,920
Speaker 1: could very clearly be wrong. Okay. Next up, Elon Musk

255
00:15:37,960 --> 00:15:40,480
Speaker 1: had some less than elegant words for advertisers who are

256
00:15:40,520 --> 00:15:44,080
Speaker 1: fleeing X, the platform formerly known as Twitter. At the

257
00:15:44,200 --> 00:15:47,200
Speaker 1: Deal Book conference, Musk had a bit of a rant

258
00:15:47,360 --> 00:15:51,400
Speaker 1: against advertisers, saying they were in effect blackmailing him. Now

259
00:15:51,440 --> 00:15:54,920
Speaker 1: as a reminder, Musk had posted and also boosted posts

260
00:15:55,040 --> 00:15:59,720
Speaker 1: containing antisemitic messaging, and many advertisers chose that moment to

261
00:15:59,760 --> 00:16:02,360
Speaker 1: say peace out to the platform. So I'm not sure

262
00:16:02,400 --> 00:16:04,720
Speaker 1: this is really a case of blackmail. To be clear,

263
00:16:05,040 --> 00:16:07,320
Speaker 1: I think this is a case of advertisers not wanting

264
00:16:07,400 --> 00:16:10,440
Speaker 1: to be complicit in the spread of hate speech because

265
00:16:10,440 --> 00:16:13,440
Speaker 1: that's a pretty bad way to protect your brand's image.

266
00:16:13,960 --> 00:16:17,600
Speaker 1: So that's not the same thing as blackmail. Facing consequences

267
00:16:17,600 --> 00:16:21,840
Speaker 1: for your actions is not the same thing as being persecuted. Anyway,

268
00:16:22,320 --> 00:16:27,000
Speaker 1: Musk told the advertisers to well, using a phrase that

269
00:16:27,040 --> 00:16:29,520
Speaker 1: would be used in the television series The Good Place

270
00:16:29,840 --> 00:16:33,680
Speaker 1: to Go, fork themselves. Then he made himself the victim

271
00:16:33,840 --> 00:16:37,520
Speaker 1: by saying that if the company dies, which seems possible,

272
00:16:37,720 --> 00:16:41,080
Speaker 1: it will be because of an advertiser boycott. I think

273
00:16:41,640 --> 00:16:45,920
Speaker 1: if x slash Twitter dies, it's because an entitled billionaire

274
00:16:45,920 --> 00:16:48,840
Speaker 1: who thinks consequences are something that happens to other people

275
00:16:49,280 --> 00:16:52,360
Speaker 1: bought the company on a whim. And remember this was

276
00:16:52,840 --> 00:16:55,840
Speaker 1: after he announced his intention to buy it, then tried

277
00:16:55,880 --> 00:16:59,040
Speaker 1: to back out of the deal almost immediately, and then

278
00:16:59,200 --> 00:17:01,400
Speaker 1: went through the deal only when he was threatened with

279
00:17:01,480 --> 00:17:07,000
Speaker 1: legal consequences I think perhaps that's the reason that x

280
00:17:07,040 --> 00:17:11,800
Speaker 1: slash Twitter dies, not that advertisers just spontaneously decided they

281
00:17:11,840 --> 00:17:15,600
Speaker 1: don't like Elon Musk. But what do I know? The

282
00:17:15,760 --> 00:17:19,639
Speaker 1: United Auto Workers labor group recently negotiated new contracts with

283
00:17:19,720 --> 00:17:23,480
Speaker 1: three major automakers here in the US. Now, as revealed

284
00:17:23,480 --> 00:17:27,479
Speaker 1: in a video recorded by UAW President Sean Fain, it

285
00:17:27,560 --> 00:17:32,440
Speaker 1: aims to unionize Tesla Motors. Considering the union secured substantial

286
00:17:32,520 --> 00:17:36,560
Speaker 1: raises for its members with those three major automakers, this

287
00:17:36,840 --> 00:17:40,400
Speaker 1: effort might actually gain some interest within Tesla. The UAW

288
00:17:40,520 --> 00:17:43,879
Speaker 1: also wants to organize workers at Mercedes Benz and at Toyota,

289
00:17:43,960 --> 00:17:47,480
Speaker 1: among other auto manufacturers. I imagine Elon Musk is not

290
00:17:47,640 --> 00:17:51,639
Speaker 1: happy about this. He recently expressed anger and frustration over

291
00:17:51,720 --> 00:17:55,440
Speaker 1: the opposition that labor groups have shown against Tesla over

292
00:17:55,480 --> 00:17:58,760
Speaker 1: in Sweden. That's been an ongoing story, and Musk has

293
00:17:58,800 --> 00:18:01,880
Speaker 1: not been quiet about his desay pleasure at the situation.

294
00:18:02,520 --> 00:18:05,720
Speaker 1: We've been seeing a growth in unions and organization over

295
00:18:05,760 --> 00:18:07,560
Speaker 1: the past few years. Have been covering it for like

296
00:18:07,800 --> 00:18:10,800
Speaker 1: three years. We'll have to wait and see if they

297
00:18:10,880 --> 00:18:14,399
Speaker 1: take hold in Tesla. There have been individual efforts at

298
00:18:14,480 --> 00:18:19,080
Speaker 1: various Tesla facilities for unionization, but this would be more

299
00:18:19,080 --> 00:18:23,440
Speaker 1: of an organization wide approach. For a quarter mile down

300
00:18:23,480 --> 00:18:27,240
Speaker 1: Fourteenth Street in Detroit, you can recharge your electric vehicle

301
00:18:27,840 --> 00:18:31,399
Speaker 1: even as you drive it. The company Electrion completed the

302
00:18:31,440 --> 00:18:35,399
Speaker 1: installation of some inductive charging coils along that stretch of

303
00:18:35,560 --> 00:18:38,719
Speaker 1: street and it will serve as a test for the technology,

304
00:18:38,800 --> 00:18:41,680
Speaker 1: so evs that are equipped with an appropriate receiver will

305
00:18:41,680 --> 00:18:43,520
Speaker 1: be able to make use of the coils. The coils

306
00:18:43,520 --> 00:18:46,960
Speaker 1: will create an electromagnetic field. Passing through the field will

307
00:18:46,960 --> 00:18:49,520
Speaker 1: induce electric charge to flow through the receiver, which then

308
00:18:49,760 --> 00:18:53,720
Speaker 1: gets directed to recharge the battery, so your car can

309
00:18:53,840 --> 00:18:56,760
Speaker 1: charge as it drives along the street. The tech also

310
00:18:56,800 --> 00:18:58,840
Speaker 1: works at the car's stationary so if you park your

311
00:18:58,920 --> 00:19:02,480
Speaker 1: vehicle along that stretch of street, it will recharge while

312
00:19:02,520 --> 00:19:04,199
Speaker 1: you do whatever it is you have to do, and

313
00:19:04,240 --> 00:19:07,320
Speaker 1: you don't have to have any plugs or cables required. Now,

314
00:19:07,359 --> 00:19:10,560
Speaker 1: this is an early test of the technology. It's really

315
00:19:10,600 --> 00:19:13,400
Speaker 1: looked as sort of a test bed for the tech.

316
00:19:13,800 --> 00:19:15,280
Speaker 1: It's not quite at the level where you're going to

317
00:19:15,320 --> 00:19:17,600
Speaker 1: see it installed in cities worldwide, but if it is

318
00:19:17,640 --> 00:19:21,080
Speaker 1: shown to be practical, we could see other municipalities around

319
00:19:21,080 --> 00:19:24,600
Speaker 1: the world invest in this technology. And that's it for

320
00:19:24,640 --> 00:19:26,760
Speaker 1: the tech news that I have for you today, Thursday,

321
00:19:26,800 --> 00:19:29,359
Speaker 1: November thirty, of twenty twenty three. I am out on

322
00:19:29,480 --> 00:19:31,919
Speaker 1: vacation next week, so we will have a few rerun

323
00:19:31,960 --> 00:19:35,800
Speaker 1: episodes play next week. We also will have a special

324
00:19:35,880 --> 00:19:39,560
Speaker 1: episode of another podcast play on Tuesday, so that you

325
00:19:39,600 --> 00:19:42,440
Speaker 1: can be introduced to a new show that I think

326
00:19:42,560 --> 00:19:46,000
Speaker 1: is pretty interesting. It takes a different look at AI

327
00:19:46,160 --> 00:19:48,760
Speaker 1: than you'll typically hear on tech Stuff, and honestly, as

328
00:19:48,880 --> 00:19:52,040
Speaker 1: much as I him and haw about artificial intelligence on

329
00:19:52,040 --> 00:19:54,760
Speaker 1: this show, I do think it's really important to hear

330
00:19:54,800 --> 00:19:57,760
Speaker 1: different points of view because I can't claim to be

331
00:19:58,840 --> 00:20:02,200
Speaker 1: the correct person. I don't think I am just right

332
00:20:02,320 --> 00:20:05,919
Speaker 1: in everything. My opinions have been shaped by various experiences.

333
00:20:06,040 --> 00:20:08,520
Speaker 1: I think it's very important to hear other points of view.

334
00:20:08,720 --> 00:20:11,600
Speaker 1: So check that out and I will talk to you

335
00:20:11,680 --> 00:20:22,199
Speaker 1: again really soon. Tech Stuff is an iHeartRadio production. For

336
00:20:22,320 --> 00:20:27,160
Speaker 1: more podcasts from iHeartRadio, visit the iHeartRadio app, Apple Podcasts,

337
00:20:27,280 --> 00:20:29,280
Speaker 1: or wherever you listen to your favorite shows.