1
00:00:04,440 --> 00:00:12,280
Speaker 1: Welcome to tech Stuff, a production from iHeartRadio. Hey there,

2
00:00:12,280 --> 00:00:16,000
Speaker 1: and welcome to tech Stuff. I'm your host, Jonathan Strickland.

3
00:00:16,120 --> 00:00:19,920
Speaker 1: I'm an executive producer with iHeartRadio. And how the tech

4
00:00:20,000 --> 00:00:23,480
Speaker 1: are you. It's time for the tech news for Thursday,

5
00:00:23,960 --> 00:00:29,720
Speaker 1: April twentieth, twenty twenty three. So it's four twenty and

6
00:00:29,840 --> 00:00:33,480
Speaker 1: fittingly enough, our first story is one where technology went

7
00:00:33,680 --> 00:00:39,400
Speaker 1: up in smoke. So first up. SpaceX's Starship launched this

8
00:00:39,560 --> 00:00:43,760
Speaker 1: morning after scrubbing the previously scheduled launch earlier this week.

9
00:00:44,320 --> 00:00:47,840
Speaker 1: A frozen valve was the reason behind the earlier cancelation.

10
00:00:48,400 --> 00:00:53,320
Speaker 1: But today the world's largest, most powerful launch vehicle in

11
00:00:53,520 --> 00:00:58,760
Speaker 1: our history of space exploration lifted up off its platform

12
00:00:58,840 --> 00:01:03,120
Speaker 1: in South Texas to hurtle toward orbit, though it did

13
00:01:03,120 --> 00:01:07,960
Speaker 1: not actually get there because around four minutes after launch,

14
00:01:08,520 --> 00:01:13,679
Speaker 1: things got spicy. The first stage engines cut off as planned,

15
00:01:14,360 --> 00:01:18,040
Speaker 1: but the second stage failed to separate from the first

16
00:01:18,120 --> 00:01:21,720
Speaker 1: stage and the launch vehicle began spinning, and then there

17
00:01:21,840 --> 00:01:25,240
Speaker 1: was a burst of flame and an explosion, with a

18
00:01:25,360 --> 00:01:30,080
Speaker 1: vehicle breaking apart in the process. Or, as SpaceX's Twitter

19
00:01:30,120 --> 00:01:33,880
Speaker 1: feed put it, Starship just experienced what we call a

20
00:01:34,040 --> 00:01:39,479
Speaker 1: rapid unscheduled disassembly. The SpaceX team stressed that the main

21
00:01:39,600 --> 00:01:43,119
Speaker 1: goal of this test was to clear the launch tower,

22
00:01:43,560 --> 00:01:46,520
Speaker 1: you know, to actually get a launch and a lift off,

23
00:01:46,600 --> 00:01:49,720
Speaker 1: and to clear the launch tower, which the starship did.

24
00:01:50,160 --> 00:01:53,320
Speaker 1: So with that in mind, the test was a success.

25
00:01:53,440 --> 00:01:56,960
Speaker 1: It's easy to say otherwise because subsequently it then exploded,

26
00:01:57,760 --> 00:02:01,000
Speaker 1: but it was clearly a disappointment that the spacecraft was

27
00:02:01,080 --> 00:02:04,880
Speaker 1: unable to deliver the payload into orbit as hoped. Regardless,

28
00:02:05,160 --> 00:02:08,640
Speaker 1: the launch vehicle has more tests in its future. Even

29
00:02:08,720 --> 00:02:12,399
Speaker 1: if the test had been completely successful, it still would

30
00:02:12,480 --> 00:02:14,960
Speaker 1: have lots more tests in the future. As to what

31
00:02:15,160 --> 00:02:17,840
Speaker 1: caused the failure, I'm sure there will be a full

32
00:02:17,880 --> 00:02:21,880
Speaker 1: accounting of it later. I saw early speculation that several

33
00:02:21,880 --> 00:02:24,359
Speaker 1: engines on one side of the first stage cut out early.

34
00:02:24,880 --> 00:02:30,880
Speaker 1: I've seen other speculation that SpaceX chose to detonate the spacecraft,

35
00:02:31,320 --> 00:02:34,040
Speaker 1: but as of the time I'm actually recording this, I

36
00:02:34,120 --> 00:02:37,000
Speaker 1: haven't seen official word, so I'm sure by the time

37
00:02:37,000 --> 00:02:39,000
Speaker 1: you listen to this that will have been cleared up.

38
00:02:39,320 --> 00:02:42,160
Speaker 1: But as a recorded that has not yet happened. Now,

39
00:02:42,200 --> 00:02:46,040
Speaker 1: he may as well get the other Musk related stories

40
00:02:46,080 --> 00:02:49,320
Speaker 1: out of the way, because Elon Musk, obviously head of SpaceX.

41
00:02:50,120 --> 00:02:55,240
Speaker 1: He appeared at an advertising conference yesterday Wednesday with a

42
00:02:55,280 --> 00:02:58,520
Speaker 1: message for the crowd that Twitter is a safe place

43
00:02:58,639 --> 00:03:01,840
Speaker 1: for advertisers to do business. Also, he doesn't plan to

44
00:03:01,880 --> 00:03:05,240
Speaker 1: restrict the types of speech that brands may want to

45
00:03:05,280 --> 00:03:09,799
Speaker 1: try and avoid, So he's saying Twitter's safe, but there's

46
00:03:09,800 --> 00:03:13,440
Speaker 1: also no guarantee that a brand's messaging won't appear alongside

47
00:03:13,800 --> 00:03:18,240
Speaker 1: objectionable speech, which probably isn't much of a selling point

48
00:03:18,400 --> 00:03:22,360
Speaker 1: for many advertisers out there. He also sent the message

49
00:03:22,639 --> 00:03:26,200
Speaker 1: don't tell us what to do. He essentially said, don't

50
00:03:26,240 --> 00:03:29,560
Speaker 1: try and dictate what Twitter should do. That's my job.

51
00:03:30,120 --> 00:03:32,919
Speaker 1: I'm not really sure if the strategy is going to work.

52
00:03:33,360 --> 00:03:36,800
Speaker 1: The company has seen some pretty drastic drops in revenue

53
00:03:36,920 --> 00:03:41,760
Speaker 1: from advertising anyway. A bunch of the regular customers with

54
00:03:42,000 --> 00:03:47,120
Speaker 1: advertising on Twitter dropped the platform entirely. A lot of

55
00:03:47,120 --> 00:03:50,200
Speaker 1: others stuck around, but they also reduced their spending on

56
00:03:50,240 --> 00:03:53,400
Speaker 1: the platform by as much as eighty percent. Of course,

57
00:03:53,880 --> 00:03:56,400
Speaker 1: some of that may have nothing to do with Twitter

58
00:03:56,440 --> 00:03:59,480
Speaker 1: as a platform. The decision might be more due to

59
00:03:59,520 --> 00:04:02,880
Speaker 1: the overall economic situation where a lot of companies have

60
00:04:03,360 --> 00:04:09,360
Speaker 1: rained in marketing expenditures. Meanwhile, Microsoft announced its Smart Campaigns

61
00:04:09,400 --> 00:04:13,040
Speaker 1: with a Multiplatform Ad Service, which is a way for

62
00:04:13,120 --> 00:04:17,400
Speaker 1: advertisers to use a tool to plan out their advertising

63
00:04:17,440 --> 00:04:21,440
Speaker 1: strategy on different platforms. They said that they're actually dropping

64
00:04:21,480 --> 00:04:25,520
Speaker 1: Twitter starting April twenty five. Other platforms will still be

65
00:04:25,640 --> 00:04:30,960
Speaker 1: available through this tool, but Twitter will not, So advertisers

66
00:04:31,000 --> 00:04:34,400
Speaker 1: who use Microsoft platform for things like ad strategy and

67
00:04:34,440 --> 00:04:37,320
Speaker 1: placement are going to have to rely on something else

68
00:04:37,400 --> 00:04:41,039
Speaker 1: with regard to Twitter. I'm sure that this will lead

69
00:04:41,040 --> 00:04:44,720
Speaker 1: to more speculation about if or when Twitter will fold.

70
00:04:45,480 --> 00:04:48,320
Speaker 1: I'm not ready to make any kind of prediction like that.

71
00:04:48,839 --> 00:04:50,760
Speaker 1: I'm not even going to predict that it will fold,

72
00:04:50,839 --> 00:04:54,560
Speaker 1: let alone when it will fold, because as I look

73
00:04:54,680 --> 00:04:58,200
Speaker 1: back over all the things that have happened with Twitter

74
00:04:58,480 --> 00:05:01,600
Speaker 1: since Musk first announced his intent to buy the company,

75
00:05:02,600 --> 00:05:05,080
Speaker 1: there is no way I could have anticipated all the

76
00:05:05,080 --> 00:05:07,720
Speaker 1: stuff that followed. So I sure as heck am not

77
00:05:07,839 --> 00:05:10,360
Speaker 1: going to try and predict the future on this one.

78
00:05:10,400 --> 00:05:14,480
Speaker 1: In Tesla News, the company issued a Q one update

79
00:05:14,600 --> 00:05:18,440
Speaker 1: letter in which Elon Musk announced that collectively, the beta

80
00:05:18,520 --> 00:05:22,120
Speaker 1: program for Tesla's full self driving service, which I feel

81
00:05:22,160 --> 00:05:25,520
Speaker 1: I need to mention, is not really full self driving

82
00:05:25,920 --> 00:05:29,720
Speaker 1: has now logged more than one hundred and fifty million miles,

83
00:05:30,160 --> 00:05:33,400
Speaker 1: and that really is an incredible achievement. You know, I

84
00:05:33,480 --> 00:05:36,839
Speaker 1: give Tesla a lot of flak, but credit where credits due.

85
00:05:36,920 --> 00:05:40,440
Speaker 1: One hundred and fifty million miles traveled is a lot. Also,

86
00:05:40,520 --> 00:05:43,640
Speaker 1: with so many miles traveled, Tesla has gathered a truly

87
00:05:43,839 --> 00:05:47,360
Speaker 1: enormous amount of data that can then be fed back

88
00:05:47,400 --> 00:05:53,320
Speaker 1: into systems to improve performance. So theoretically, with more miles traveled,

89
00:05:53,560 --> 00:05:56,680
Speaker 1: the system gets more reliable and safer. You can kind

90
00:05:56,680 --> 00:05:59,720
Speaker 1: of think of every car that has full self driving

91
00:05:59,720 --> 00:06:03,680
Speaker 1: act as being kind of like a student driver, only

92
00:06:03,960 --> 00:06:07,400
Speaker 1: these students all share the same brain. So as one

93
00:06:07,600 --> 00:06:11,240
Speaker 1: car learns a lesson, the benefits of that lesson go

94
00:06:11,360 --> 00:06:15,800
Speaker 1: to the entire fleet of cars. Now, I say theoretically

95
00:06:16,279 --> 00:06:18,560
Speaker 1: because there are actually other factors we have to take

96
00:06:18,560 --> 00:06:22,359
Speaker 1: into consideration. So, for example, Tesla has made changes to

97
00:06:22,440 --> 00:06:26,599
Speaker 1: the hardware in the construction of their vehicles, such as

98
00:06:26,680 --> 00:06:31,440
Speaker 1: removing ultrasonic sensors in favor of optical ones, and these

99
00:06:31,440 --> 00:06:35,400
Speaker 1: hardware changes also affect performance. So it's not just the

100
00:06:35,440 --> 00:06:40,119
Speaker 1: brains of these vehicles, it's the actual sensors as well.

101
00:06:40,600 --> 00:06:42,840
Speaker 1: It really doesn't matter how smart your vehicle is if

102
00:06:42,880 --> 00:06:47,360
Speaker 1: it can't sense an obstacle for example. Still, it is

103
00:06:47,800 --> 00:06:53,440
Speaker 1: a legitimately impressive achievement. That being said, Tesla's investors are

104
00:06:53,520 --> 00:06:56,640
Speaker 1: not so happy with the company right now. Also in

105
00:06:56,640 --> 00:06:59,520
Speaker 1: that letter was the revelation that the company's gross margins

106
00:06:59,560 --> 00:07:03,160
Speaker 1: are below twenty percent. The net income for the company

107
00:07:03,200 --> 00:07:06,000
Speaker 1: is two point nine billion dollars. That's a big old

108
00:07:06,080 --> 00:07:08,919
Speaker 1: chunk of change. But this time last year, the company's

109
00:07:08,960 --> 00:07:12,360
Speaker 1: income was seven hundred million dollars more than that, so

110
00:07:12,920 --> 00:07:15,960
Speaker 1: it is a drop in year over year net income.

111
00:07:16,600 --> 00:07:20,400
Speaker 1: Tesla has cut vehicle prices several times since the start

112
00:07:20,480 --> 00:07:23,440
Speaker 1: of the year because demand has been low, so in

113
00:07:23,520 --> 00:07:27,800
Speaker 1: order to try and you know, stimulate demand, they've cut

114
00:07:27,800 --> 00:07:30,320
Speaker 1: some prices. And my guess is the tough economy is

115
00:07:30,360 --> 00:07:33,880
Speaker 1: really playing a part in this, although another big part

116
00:07:34,320 --> 00:07:36,720
Speaker 1: is that other car companies are starting to finally make

117
00:07:36,800 --> 00:07:40,040
Speaker 1: up ground on Tesla, which for years had the advantage

118
00:07:40,040 --> 00:07:43,560
Speaker 1: of being all in on electric vehicles and most other

119
00:07:43,640 --> 00:07:46,800
Speaker 1: car companies were just kind of dipping their toe in

120
00:07:46,880 --> 00:07:49,360
Speaker 1: the water at the time. Well that's starting to change

121
00:07:49,400 --> 00:07:52,640
Speaker 1: now and now Tesla's facing greater competition in the electric

122
00:07:52,680 --> 00:07:56,480
Speaker 1: vehicle market. So Musk tried to get ahead of a

123
00:07:56,520 --> 00:08:01,120
Speaker 1: negative investor reaction. He said, quote, you're taking a view

124
00:08:01,160 --> 00:08:04,720
Speaker 1: that pushing for higher volumes and a larger fleet is

125
00:08:04,760 --> 00:08:09,080
Speaker 1: the right choice here versus a lower volume and higher

126
00:08:09,120 --> 00:08:13,120
Speaker 1: margin end quote. And honestly, I'm not sure he had

127
00:08:13,200 --> 00:08:15,840
Speaker 1: much other choice, because if not enough people are buying

128
00:08:15,880 --> 00:08:19,440
Speaker 1: Tesla's right now after several price cuts, they sure as

129
00:08:19,480 --> 00:08:23,400
Speaker 1: heck wouldn't be buying cars there were priced even higher.

130
00:08:23,920 --> 00:08:26,760
Speaker 1: Investors still weren't happy, though, and the stock price is

131
00:08:26,800 --> 00:08:31,440
Speaker 1: down by around six to seven percent six ' two seven.

132
00:08:32,320 --> 00:08:34,439
Speaker 1: As I was working on this episode, it was hovering

133
00:08:34,679 --> 00:08:38,520
Speaker 1: just under a seven percent drop from the day before. Okay,

134
00:08:38,640 --> 00:08:41,839
Speaker 1: now it's a high time, everyone's favorite time of the news,

135
00:08:42,559 --> 00:08:45,240
Speaker 1: and first up is a story I found more than

136
00:08:45,480 --> 00:08:49,640
Speaker 1: a little upsetting. I mean, like this one genuinely disturbed me.

137
00:08:49,960 --> 00:08:53,160
Speaker 1: All right, First, let me give some backstory. There's a

138
00:08:53,200 --> 00:08:56,960
Speaker 1: Formula one racer named Michael Schumbacher who was in a

139
00:08:57,160 --> 00:09:01,840
Speaker 1: terrible accident while skiing, not racing, and this happened a

140
00:09:01,920 --> 00:09:05,640
Speaker 1: decade ago, back in twenty thirteen. He was severely injured

141
00:09:05,679 --> 00:09:08,880
Speaker 1: in that accident. He received a brain injury in the process,

142
00:09:09,440 --> 00:09:12,520
Speaker 1: and since then he's been out of the limelight. His

143
00:09:12,600 --> 00:09:15,760
Speaker 1: family has stressed the need for privacy. They continue to

144
00:09:15,880 --> 00:09:20,000
Speaker 1: help him day by day. And that story alone is

145
00:09:20,000 --> 00:09:23,400
Speaker 1: a very tough one. Right to see someone have that

146
00:09:23,559 --> 00:09:27,719
Speaker 1: kind of injury, that's just terrible. But then we get

147
00:09:27,760 --> 00:09:30,440
Speaker 1: to something that is really hard to believe. A German

148
00:09:30,520 --> 00:09:37,240
Speaker 1: weekly magazine called De Actual was running a supposed interview

149
00:09:37,480 --> 00:09:42,640
Speaker 1: with Schumacher. Only it wasn't an interview with Schumacher, he

150
00:09:42,720 --> 00:09:47,760
Speaker 1: has not given any interviews since this accident in twenty thirteen. Rather,

151
00:09:48,400 --> 00:09:51,280
Speaker 1: it was an AI agent that had been instructed to

152
00:09:51,360 --> 00:09:56,800
Speaker 1: respond to questions as if it were Michael Schumacher. So

153
00:09:57,280 --> 00:10:01,360
Speaker 1: this weekly magazine published a quote unquote interview with an

154
00:10:01,400 --> 00:10:05,960
Speaker 1: AI chatbot impersonating an athlete who is still coping with

155
00:10:06,000 --> 00:10:08,719
Speaker 1: the effects of a brain injury. So to call it

156
00:10:09,360 --> 00:10:14,080
Speaker 1: tasteless and unethical seems to be a gross understatement, And

157
00:10:14,120 --> 00:10:15,800
Speaker 1: it kind of harkens back to what I was saying

158
00:10:15,840 --> 00:10:20,600
Speaker 1: earlier this week about concepts like right to identity and

159
00:10:20,760 --> 00:10:25,280
Speaker 1: right to personality. I mean, imagine if some news outlet

160
00:10:25,400 --> 00:10:28,960
Speaker 1: there out there ran a supposed interview with you, but

161
00:10:29,040 --> 00:10:31,080
Speaker 1: it wasn't with you. It was with a chatbot that

162
00:10:31,240 --> 00:10:35,240
Speaker 1: was told to impersonate you, and then it's presented to

163
00:10:35,280 --> 00:10:38,000
Speaker 1: people as if it's a legitimate interview. That would be

164
00:10:38,720 --> 00:10:43,640
Speaker 1: an incredible violation. Schumacher's family are pursuing legal action against

165
00:10:43,679 --> 00:10:47,199
Speaker 1: the magazine and y'all. A lot of the concerns around

166
00:10:47,280 --> 00:10:51,280
Speaker 1: AI may be a little premature, Like the worry that

167
00:10:51,320 --> 00:10:54,320
Speaker 1: AI is going to doom the human race might be

168
00:10:54,360 --> 00:10:57,960
Speaker 1: a bit of an overreaction as things are right now.

169
00:10:58,120 --> 00:11:01,160
Speaker 1: But when you see media outlets willing to go to

170
00:11:01,200 --> 00:11:04,040
Speaker 1: these kinds of links to generate a story, you realize

171
00:11:04,080 --> 00:11:07,800
Speaker 1: that AI could be dangerous. Sure, it is dangerous in

172
00:11:07,840 --> 00:11:12,720
Speaker 1: several ways, but human beings can be downright diabolical. Okay,

173
00:11:12,880 --> 00:11:15,000
Speaker 1: we've got a lot more stories to go to, but

174
00:11:15,120 --> 00:11:27,800
Speaker 1: first let's take a quick break. Okay, we're back. Let's

175
00:11:27,840 --> 00:11:31,880
Speaker 1: talk some more about AI. So, Reddit's CEO Steve Huffman

176
00:11:32,400 --> 00:11:35,240
Speaker 1: has a message to AI companies out there. If they

177
00:11:35,280 --> 00:11:38,560
Speaker 1: want to scrape content from Reddit as part of creating

178
00:11:38,800 --> 00:11:42,640
Speaker 1: large language models and training AI agents, they're going to

179
00:11:42,720 --> 00:11:45,080
Speaker 1: have to pay for that privilege. He spoke with The

180
00:11:45,120 --> 00:11:47,240
Speaker 1: New York Times and said Reddit plans to make an

181
00:11:47,280 --> 00:11:52,240
Speaker 1: exception in its API, its Application programming interface that if

182
00:11:52,280 --> 00:11:56,600
Speaker 1: an AI company wants to access this tool, they'll have

183
00:11:56,679 --> 00:11:59,440
Speaker 1: to pay first. Now, if you're a developer who needs

184
00:11:59,480 --> 00:12:03,959
Speaker 1: access to that API for something other than training artificial intelligence,

185
00:12:04,360 --> 00:12:07,400
Speaker 1: good news, no charge for you. You get to use

186
00:12:07,440 --> 00:12:10,880
Speaker 1: the tool. As intended. As for how much AI companies

187
00:12:10,920 --> 00:12:14,600
Speaker 1: will have to pay to get access, that really hasn't

188
00:12:14,640 --> 00:12:17,760
Speaker 1: been hashed out yet. Reddit has already served as a

189
00:12:17,840 --> 00:12:21,520
Speaker 1: treasure trove of human centric data for companies like Google

190
00:12:21,640 --> 00:12:26,160
Speaker 1: and OpenAI while they were developing their respective artificial intelligence

191
00:12:26,240 --> 00:12:30,200
Speaker 1: large language models. Personally, I like this move by Reddit.

192
00:12:30,440 --> 00:12:33,960
Speaker 1: I mean, we all know that information has value. Now

193
00:12:34,040 --> 00:12:36,400
Speaker 1: if only the Reddit users could get a cut of

194
00:12:36,440 --> 00:12:38,960
Speaker 1: the action. But that's probably taking things a little too far.

195
00:12:39,840 --> 00:12:42,480
Speaker 1: In a kind of similar story, Google is leading the

196
00:12:42,559 --> 00:12:46,720
Speaker 1: charge in petitioning the Australian government to allow AI projects

197
00:12:46,760 --> 00:12:53,000
Speaker 1: to crawl Australian websites and to be used in Australian implementations.

198
00:12:53,080 --> 00:12:58,120
Speaker 1: So currently Australia's copyright laws are potentially a barrier, creating

199
00:12:58,520 --> 00:13:02,320
Speaker 1: pesky legal restrictions on what Google and other companies like

200
00:13:02,360 --> 00:13:05,959
Speaker 1: open ai would consider to be information that's free for

201
00:13:06,040 --> 00:13:10,560
Speaker 1: the taking and applications that should be fully open to them.

202
00:13:11,120 --> 00:13:14,840
Speaker 1: Google framed its request by saying Australia could really be

203
00:13:14,920 --> 00:13:19,000
Speaker 1: missing out, with one spokesperson stating quote, the lack of

204
00:13:19,080 --> 00:13:23,840
Speaker 1: such copyright flexibilities means that investment in and development of

205
00:13:24,160 --> 00:13:28,600
Speaker 1: AI and machine learning technologies is happening and will continue

206
00:13:28,600 --> 00:13:32,760
Speaker 1: to happen overseas end quote. That's according to the Guardian.

207
00:13:33,240 --> 00:13:35,680
Speaker 1: So Google essentially is saying, hey, if you don't relax

208
00:13:35,720 --> 00:13:38,520
Speaker 1: those laws, my dude, you're going to have to import

209
00:13:38,600 --> 00:13:41,440
Speaker 1: all those useful AI tools from somewhere else, because ain't

210
00:13:41,440 --> 00:13:42,840
Speaker 1: no one going to be here to do it in

211
00:13:42,880 --> 00:13:46,240
Speaker 1: Australia because your laws are whack. Now. I have no

212
00:13:46,360 --> 00:13:49,719
Speaker 1: clue if Google's approach will be effective, particularly in an

213
00:13:49,800 --> 00:13:53,679
Speaker 1: environment where there's an increased wariness surrounding AI and its

214
00:13:53,679 --> 00:13:58,080
Speaker 1: development in the first place. I think the tide in

215
00:13:58,559 --> 00:14:02,240
Speaker 1: public opinion on a I is starting to turn sour.

216
00:14:02,800 --> 00:14:06,280
Speaker 1: Bloomberg has published an article titled Google's rush to win

217
00:14:06,480 --> 00:14:10,480
Speaker 1: in AI led to ethical lapses employees say, and the

218
00:14:10,640 --> 00:14:15,640
Speaker 1: article sites Google staff using language like cringeworthy or calling

219
00:14:15,800 --> 00:14:20,680
Speaker 1: its aichatbot a pathological liar. They were using these words

220
00:14:20,720 --> 00:14:24,080
Speaker 1: to describe Google Bard while it was still in development,

221
00:14:24,120 --> 00:14:27,600
Speaker 1: before it had been pushed out into a beta program.

222
00:14:28,000 --> 00:14:31,160
Speaker 1: They said the tool would frequently give unreliable or sometimes

223
00:14:31,160 --> 00:14:35,320
Speaker 1: outright dangerous information, while doing so in an authoritative voice,

224
00:14:35,320 --> 00:14:37,640
Speaker 1: so it sounds like it knows what's talking about. So

225
00:14:37,760 --> 00:14:41,440
Speaker 1: one example given in the article is about scuba diving

226
00:14:41,480 --> 00:14:45,240
Speaker 1: procedures and that supposedly Barred gave answers that would quote

227
00:14:45,800 --> 00:14:49,320
Speaker 1: likely result in serious injury or death end quote. So

228
00:14:49,360 --> 00:14:53,280
Speaker 1: that's really not good. This was during the testing phase,

229
00:14:53,320 --> 00:14:55,400
Speaker 1: as I said, for Bard, and my guess is that

230
00:14:55,480 --> 00:14:59,120
Speaker 1: Google had planned to keep Barred under wraps and continue

231
00:14:59,200 --> 00:15:02,200
Speaker 1: to work on it within development for much much longer.

232
00:15:02,760 --> 00:15:06,040
Speaker 1: But Open AI's release of chat GPT late last year

233
00:15:06,200 --> 00:15:10,280
Speaker 1: convinced Google's leadership to try the old move fast and

234
00:15:10,440 --> 00:15:14,160
Speaker 1: break things approach, or else the company would risk getting

235
00:15:14,240 --> 00:15:17,760
Speaker 1: left behind in the AI chatbot wars. And you know,

236
00:15:18,440 --> 00:15:21,880
Speaker 1: Google really hates it if it is not the dominant

237
00:15:21,880 --> 00:15:24,560
Speaker 1: player in whichever markets it's competing in. It likes to

238
00:15:24,600 --> 00:15:29,240
Speaker 1: have that comfortable sixty to ninety percent market share. So

239
00:15:29,320 --> 00:15:32,160
Speaker 1: the article reads kind of as a warning about AI

240
00:15:32,240 --> 00:15:37,240
Speaker 1: in general, that this push to create a snazzy tool

241
00:15:37,640 --> 00:15:41,760
Speaker 1: could result in harmful consequences, and that the resources required

242
00:15:41,760 --> 00:15:45,720
Speaker 1: to make certain that the development and deployment is following

243
00:15:45,800 --> 00:15:50,160
Speaker 1: really good, strong ethical guidelines those just aren't in place

244
00:15:50,360 --> 00:15:54,560
Speaker 1: or they're being ignored. Google reps told Bloomberg that quote

245
00:15:54,640 --> 00:15:58,360
Speaker 1: responsible AI remains a top priority at the company end quote. Now,

246
00:15:58,440 --> 00:16:01,360
Speaker 1: in my opinion, company's across the board have been far

247
00:16:01,440 --> 00:16:05,600
Speaker 1: too aggressive with these tools. Open AI's leadership has even

248
00:16:05,720 --> 00:16:09,960
Speaker 1: warned media outlets about its own chat pots, saying it's

249
00:16:09,960 --> 00:16:14,080
Speaker 1: not reliable. But by that time the cat was already

250
00:16:14,120 --> 00:16:16,800
Speaker 1: out of the artificial bag and it was too late

251
00:16:16,840 --> 00:16:21,280
Speaker 1: to issue cautionary messages Anyway. The Bloomberg article is well

252
00:16:21,280 --> 00:16:24,880
Speaker 1: worth a read and goes into much greater detail about

253
00:16:24,880 --> 00:16:27,640
Speaker 1: the conditions that led Google to Rushbard in to a

254
00:16:27,680 --> 00:16:33,440
Speaker 1: beta test, arguably prematurely. Yesterday, Meta held another round of layoffs,

255
00:16:33,600 --> 00:16:36,000
Speaker 1: one of three batches the company plans to spread out

256
00:16:36,040 --> 00:16:38,560
Speaker 1: across the next few months, which will ultimately see another

257
00:16:38,640 --> 00:16:42,280
Speaker 1: ten thousand Meta employees cut. This is after Meta had

258
00:16:42,320 --> 00:16:45,600
Speaker 1: already cut eleven thousand jobs late last year. This time,

259
00:16:45,640 --> 00:16:48,320
Speaker 1: the layoffs affected engineers and tech teams, and as you

260
00:16:48,400 --> 00:16:51,560
Speaker 1: might imagine, the cuts have delivered another blow to employee

261
00:16:51,560 --> 00:16:55,000
Speaker 1: morale at Meta. I mentioned in a previous news episode

262
00:16:55,000 --> 00:16:58,240
Speaker 1: that some Meta employees have become discouraged over recent months

263
00:16:58,280 --> 00:17:03,520
Speaker 1: due to several reasons. First, layoffs are demoralizing. To the

264
00:17:03,560 --> 00:17:06,520
Speaker 1: push to return to office has created a hardship on

265
00:17:06,600 --> 00:17:10,359
Speaker 1: a lot of employees. And see that hardship doesn't appear

266
00:17:10,359 --> 00:17:12,919
Speaker 1: to be shared by Meta's top executive leadership because a

267
00:17:12,920 --> 00:17:15,640
Speaker 1: lot of them seem to be absent from Meta's HQ.

268
00:17:15,840 --> 00:17:18,160
Speaker 1: Some of them have moved thousands of miles away while

269
00:17:18,160 --> 00:17:21,200
Speaker 1: still serving as executives, and yet the employees are forced

270
00:17:21,240 --> 00:17:24,920
Speaker 1: to go in. So yeah, news continues to be rough

271
00:17:25,240 --> 00:17:28,320
Speaker 1: for the staff of Meta. By the way, if you

272
00:17:28,359 --> 00:17:31,639
Speaker 1: were a Facebook user with an account any time between

273
00:17:31,720 --> 00:17:35,119
Speaker 1: May two thousand and seven and December twenty twenty two,

274
00:17:35,600 --> 00:17:39,000
Speaker 1: you can claim your share of a huge class action

275
00:17:39,200 --> 00:17:43,240
Speaker 1: lawsuit settlement. Now this has to do with Cambridge Analytica,

276
00:17:43,680 --> 00:17:46,280
Speaker 1: that scandal that rocked the tech world several years ago

277
00:17:46,800 --> 00:17:49,080
Speaker 1: and we still talk about it today. So in case

278
00:17:49,119 --> 00:17:52,760
Speaker 1: you don't remember that whole kerfuffle, a company that catered

279
00:17:52,800 --> 00:17:57,560
Speaker 1: to political campaigns, a data analytics company leveraged a survey

280
00:17:57,680 --> 00:18:00,480
Speaker 1: app to gather a huge amount of information about Facebook

281
00:18:00,560 --> 00:18:04,240
Speaker 1: users without their consent. The app was taking advantage of

282
00:18:04,400 --> 00:18:08,160
Speaker 1: large gaps in Facebook's API, gaps that the company would

283
00:18:08,160 --> 00:18:12,320
Speaker 1: subsequently close, but the damage had already been done anyway.

284
00:18:12,520 --> 00:18:14,879
Speaker 1: There were a lot of legal proceedings that followed in

285
00:18:14,920 --> 00:18:18,920
Speaker 1: the discovery of this transgression, including this class action lawsuit

286
00:18:18,920 --> 00:18:21,800
Speaker 1: that I mentioned, and the settlement was for a whopping

287
00:18:21,840 --> 00:18:25,960
Speaker 1: seven hundred and twenty five million dollars. Though Meta was

288
00:18:26,040 --> 00:18:28,000
Speaker 1: let off the hook in a sense because the company

289
00:18:28,000 --> 00:18:30,760
Speaker 1: didn't actually have to admit any wrongdoing as part of

290
00:18:30,760 --> 00:18:34,159
Speaker 1: this settlement. But then I think it's kind of hard

291
00:18:34,320 --> 00:18:37,119
Speaker 1: to say with a straight face, Hey, I didn't do

292
00:18:37,160 --> 00:18:39,320
Speaker 1: anything wrong and no one was hurt. I'm just paying

293
00:18:39,400 --> 00:18:42,760
Speaker 1: seven hundred and twenty five million dollars out of the

294
00:18:42,800 --> 00:18:45,600
Speaker 1: kindness of my heart. Anyway, the settlement itself is actually

295
00:18:45,600 --> 00:18:48,280
Speaker 1: old news. The company reached an agreement late last year.

296
00:18:48,359 --> 00:18:52,240
Speaker 1: The news now is that you Facebook users out there

297
00:18:52,240 --> 00:18:55,240
Speaker 1: who had an account that was active during those times,

298
00:18:56,000 --> 00:18:58,680
Speaker 1: you can actually submit a claim, and you have until

299
00:18:58,680 --> 00:19:01,119
Speaker 1: August twenty fifth to do so if you want to

300
00:19:01,160 --> 00:19:07,159
Speaker 1: do it online. The website is Facebook User Privacysettlement dot

301
00:19:07,200 --> 00:19:15,160
Speaker 1: com slash hashtag, submit, dash claim. Finally, mark September twenty ninth,

302
00:19:15,160 --> 00:19:18,520
Speaker 1: twenty twenty three, ondred calendars. That's when Netflix will officially

303
00:19:18,560 --> 00:19:21,639
Speaker 1: stop sending DVDs out by mail. For some of y'all,

304
00:19:21,760 --> 00:19:24,119
Speaker 1: what I said might actually sound strange, but not that

305
00:19:24,240 --> 00:19:27,840
Speaker 1: long ago. Netflix's business model was really just renting films

306
00:19:27,880 --> 00:19:31,119
Speaker 1: out through the mail. You would sign up for the service,

307
00:19:31,160 --> 00:19:33,240
Speaker 1: you would make a watch list of films and shows

308
00:19:33,280 --> 00:19:35,720
Speaker 1: you wanted to see, and then, based on what was available,

309
00:19:35,800 --> 00:19:38,800
Speaker 1: Netflix would mail you the DVD and then you would

310
00:19:38,800 --> 00:19:40,240
Speaker 1: watch it and you would put it back in its

311
00:19:40,280 --> 00:19:42,000
Speaker 1: envelope and drop it off in the mail and wait

312
00:19:42,040 --> 00:19:44,359
Speaker 1: for your next one. Or if you were like just

313
00:19:44,400 --> 00:19:46,720
Speaker 1: about everyone I know had the service, you would do

314
00:19:46,800 --> 00:19:49,320
Speaker 1: this a couple times, and then on your third DVD

315
00:19:49,640 --> 00:19:51,760
Speaker 1: you just have it sit on the coffee table in

316
00:19:51,800 --> 00:19:55,159
Speaker 1: the little red envelope, unopened and unwatched, and then you

317
00:19:55,160 --> 00:19:57,439
Speaker 1: would just continue paying for the privilege of giving this

318
00:19:57,520 --> 00:20:00,640
Speaker 1: DVD a temporary home. Or in fact, you might ultimately

319
00:20:00,680 --> 00:20:03,800
Speaker 1: get charged for the DVD because he kept it so

320
00:20:03,880 --> 00:20:07,480
Speaker 1: ding dang long. But anyway, Netflix's main business was all

321
00:20:07,520 --> 00:20:10,760
Speaker 1: in this male centric model. The company was founded back

322
00:20:10,760 --> 00:20:13,639
Speaker 1: in nineteen ninety seven, and it didn't start streaming until

323
00:20:13,720 --> 00:20:17,400
Speaker 1: two thousand and seven, so for a decade and some change,

324
00:20:17,720 --> 00:20:20,560
Speaker 1: that's just how folks use Netflix. Now we're at the

325
00:20:20,600 --> 00:20:23,160
Speaker 1: end of an era, and it's funny because I've actually

326
00:20:23,240 --> 00:20:26,159
Speaker 1: kind of gotten back into buying DVDs and Blu ray discs.

327
00:20:26,560 --> 00:20:30,520
Speaker 1: I got tired of titles disappearing off various services, sometimes

328
00:20:30,560 --> 00:20:33,560
Speaker 1: popping up on other services that I don't subscribe to,

329
00:20:33,880 --> 00:20:37,720
Speaker 1: So now I'm back to purchasing physical media. I guess

330
00:20:37,720 --> 00:20:41,560
Speaker 1: I'm just a human being, lost in time and lost

331
00:20:41,600 --> 00:20:46,600
Speaker 1: in space and in meaning. Hope you're all well, that's

332
00:20:46,640 --> 00:20:50,800
Speaker 1: it for this episode. I'll talk to you again really soon.

333
00:20:57,240 --> 00:21:01,879
Speaker 1: Tech Stuff is an iHeartRadio production. For more podcasts from iHeartRadio,

334
00:21:02,200 --> 00:21:05,920
Speaker 1: visit the iHeartRadio app, Apple Podcasts, or wherever you listen

335
00:21:05,960 --> 00:21:07,000
Speaker 1: to your favorite shows.