WEBVTT 1 00:00:01.720 --> 00:00:06.639 This is Spacetime Series twenty six, Episode fifty six, for broadcast on the 2 00:00:06.679 --> 00:00:12.759 tenth of May twenty twenty three. Coming up on Spacetime, studying powerful stellar 3 00:00:12.919 --> 00:00:18.879 eruptions called super flares. Europe announces plans to build a nuclear powered rocket, 4 00:00:19.359 --> 00:00:23.160 and we look at the work done by the European space agencies. Are matter 5 00:00:23.280 --> 00:00:29.960 of satellites undertaking air quality missions. All that and more coming up on Spacetime. 6 00:00:31.559 --> 00:00:53.960 Welcome to Spacetime with Stewart Garry. Astronomers using a combination of space and 7 00:00:54.079 --> 00:00:59.119 ground based observations have captured one of the most complete pictures yet of a super 8 00:00:59.159 --> 00:01:03.640 flare erupt from a distant star. The findings are reported in the Estrophysical Journal, 9 00:01:03.760 --> 00:01:10.239 describes a very massive, high velocity prominence eruption lasting out from the binary 10 00:01:10.319 --> 00:01:15.079 star V thirteen fifty five arianis located some four hundred light years away in the 11 00:01:15.159 --> 00:01:21.760 constellation O Ryan. The study is important because many stars are observed releasing super 12 00:01:21.760 --> 00:01:26.079 flares, which can be over ten times larger than the largest solar flares ever 13 00:01:26.120 --> 00:01:30.200 seen erupting from the Sun, and it raises the possibility of whether or not 14 00:01:30.319 --> 00:01:36.879 the Sun could release such flares as well. Hot ironized gas released by the 15 00:01:36.959 --> 00:01:42.000 Sun. Solar flares can have a major influence on the environment around the Earth, 16 00:01:42.239 --> 00:01:47.359 referred to a space whether a solar flare or chronal mass ejection, can 17 00:01:47.439 --> 00:01:53.000 cause Earth's atmosphere to wobble like a jellyfish. It triggers spectacular Aurora light displays, 18 00:01:53.120 --> 00:01:57.719 the Northern and Southern lights, the Aurora borealis and Aurora Estrallas, which 19 00:01:57.719 --> 00:02:01.480 are the displays we've been had over the past few weeks. But it can 20 00:02:01.519 --> 00:02:07.799 also damage spacecraft, frying their electrical circuits and cause them to use up more 21 00:02:07.879 --> 00:02:13.800 fuel in order to maintain operational orbits as the atmosphere around them expands and contracts, 22 00:02:13.919 --> 00:02:20.360 increasing atmospheric drag and consequently orbital decay. The radiation from these flares can 23 00:02:20.400 --> 00:02:25.199 also damage life, affecting astronauts aboard the International Space Station and on deep space 24 00:02:25.240 --> 00:02:31.280 missions. It can disrupt navigation and communication systems, and even affect power grids 25 00:02:31.319 --> 00:02:38.080 on Earth, causing widespread blackouts by overloading circuits. And they're just the big 26 00:02:38.080 --> 00:02:42.199 ones from the Sun we know about. Could out Sun produce flares ten times 27 00:02:42.280 --> 00:02:47.039 bigger, flares big enough to irradiate the planet and destroy life. The star 28 00:02:47.280 --> 00:02:53.840 V thirteen fifty five Orionus is known to frequently release large scales super flares. 29 00:02:53.520 --> 00:02:59.719 However, the details of how super flares and prominence eruptions on stars occur is 30 00:02:59.719 --> 00:03:04.520 still claire. Now, a team of scientists led by shouldn't inu from Kiota 31 00:03:04.599 --> 00:03:09.120 University have used the three point eight meter semi telescope in Japan and NASA's Transiting 32 00:03:09.159 --> 00:03:15.840 Exoplanet Survey satellite tests to monitor the V thirteen fifty five Ranus binary system, 33 00:03:15.879 --> 00:03:23.479 and the team succeeded in capturing a super flare with continuous high temporal resolution observations. 34 00:03:23.080 --> 00:03:29.000 The data analyzes shows that the super flare originated with a phenomenon known as 35 00:03:29.000 --> 00:03:32.879 a prominence eruption. Now, calculating the velocity of the eruption requires making some 36 00:03:34.120 --> 00:03:39.639 assumptions about aspects that aren't directly observable, but even the most conservative estimates far 37 00:03:39.840 --> 00:03:46.120 exceeded three hundred forty seven kilometer per second escape velocity of the star, and 38 00:03:46.240 --> 00:03:51.479 that indicates that the prominence eruption was capable of breaking free of the star's gravity 39 00:03:51.599 --> 00:03:55.400 and developing into a chronal mass ejection. The prominence eruption was also one of 40 00:03:55.439 --> 00:04:00.599 the most massive ever observed carrying trillions of tons of men material. Over All, 41 00:04:00.639 --> 00:04:05.599 the observations of providing astronomers with a better idea of how super flares and 42 00:04:05.840 --> 00:04:13.800 still a prominence eruptions can occur. This is spacetime still to come. Europe 43 00:04:13.800 --> 00:04:17.519 announcers plans to follow in the footsteps of the United States and develop a nuclear 44 00:04:17.639 --> 00:04:23.480 powered rocket. And we look at the work done by the European space agencies 45 00:04:23.560 --> 00:04:30.079 constellation of airmachoring spacecraft, the Copernicus Sentinel program, all that and more still 46 00:04:30.120 --> 00:04:49.560 to come on spacetime. The European Space Agency has announced plans to follow in 47 00:04:49.600 --> 00:04:57.199 the footsteps of NASA and develop nuclear powered rockets for deep space exploration. Spacecraft 48 00:04:57.319 --> 00:05:02.319 currently use a variety of chemical rockets or electric ion propulsion systems, usually powered 49 00:05:02.360 --> 00:05:08.600 by solar arrays, and of course there are also lots of solar serial experiments 50 00:05:08.639 --> 00:05:13.600 continuing as a third potential option for the future. But NASCY began looking at 51 00:05:13.600 --> 00:05:17.240 the idea of nuclear powered rockets back in the nineteen sixties under what was called 52 00:05:17.279 --> 00:05:23.319 projector Ryan, but it was dropped due to funding constraints and the Nuclear Non 53 00:05:23.360 --> 00:05:29.120 Proliferation Treaty. The problem is conventional space propulsion methods and now reaching their physical 54 00:05:29.199 --> 00:05:33.920 limits. One of the original ideas involved nuclear impulse detonation systems, where small 55 00:05:33.959 --> 00:05:39.399 atomic bombs would be ejected behind the rear of the spacecraft, ignited, and 56 00:05:39.480 --> 00:05:45.319 the shockwaves from the nuclear blast then used to push the spacecraft forward. Likely, 57 00:05:45.360 --> 00:05:49.079 they're no longer being considered as an option. Instead, scientists are looking 58 00:05:49.079 --> 00:05:56.319 at nuclear based electrical propulsion systems, like the electric ion propulsion systems on which 59 00:05:56.319 --> 00:06:01.480 they're based. Nuclear electric propulsion systems used propel much more efficiently than chemical rockets, 60 00:06:01.639 --> 00:06:06.279 but provide far lower amounts of thrust. They would use a nuclear reactor 61 00:06:06.360 --> 00:06:13.079 to generate electricity that positively charges gas propellants like zeon or crypton, pushing the 62 00:06:13.120 --> 00:06:17.519 ions through a thruster, which then drives the spacecraft forward. Conventional iron propulsion 63 00:06:17.560 --> 00:06:21.879 systems do the same thing, but use solar erase to generate the electricity. 64 00:06:23.600 --> 00:06:29.319 Using low thrust. Efficiently, nuclear electric propulsion systems could accelerate spacecraft for extended 65 00:06:29.399 --> 00:06:32.560 periods of time, propelling a vehicle and emission to say Mars, in just 66 00:06:32.600 --> 00:06:36.120 a fraction of the time and for a fraction of the propellant of a high 67 00:06:36.160 --> 00:06:43.519 thrust chemical system. Another idea being examined is nuclear, a thermal propulsion technology. 68 00:06:44.160 --> 00:06:48.199 It provides high thrust and twice the propellant efficiency of any chemical rocket system 69 00:06:48.279 --> 00:06:54.279 yet developed. It works by transferring heat from a nuclear reactor to a liquid 70 00:06:54.319 --> 00:06:58.560 propellant, such as liquified hydrogen, which is being pumped through the reactor core. 71 00:06:59.199 --> 00:07:02.439 This rapidly heats up, expanding to a gas which is then forced out 72 00:07:02.439 --> 00:07:08.519 of the rocket nozzle, creating high thrust and accelerating the spacecraft rapidly forward. 73 00:07:09.199 --> 00:07:14.079 The advantage of nuclear thermal propulsion is that it cannot only shorten travel times, 74 00:07:14.120 --> 00:07:18.199 but also deliver vi graded payloads. For example, a trip to Mars could 75 00:07:18.199 --> 00:07:23.360 be reduced by a quarter, cutting a six month one way trip down to 76 00:07:23.399 --> 00:07:27.199 a list than five months. And of course, reducing flight times would be 77 00:07:27.360 --> 00:07:30.360 especially important for a man mission, as the short travel times would reduce the 78 00:07:30.399 --> 00:07:34.920 amount of exposure to harmful cosmic radiation that a crew would be exposed to. 79 00:07:35.800 --> 00:07:42.800 NASA began its latest nuclear propulsion research program last year in collaboration with the Pentagon's 80 00:07:42.879 --> 00:07:47.800 Defense Advanced Research Projects Agency DAPPER. They plan on developing a nuclear thermal propulsion 81 00:07:47.879 --> 00:07:54.120 engine by twenty twenty seven. Esabes. It could have its system operational by 82 00:07:54.160 --> 00:08:00.600 twenty thirty five. This space time still to come. Take a look at 83 00:08:00.639 --> 00:08:05.399 the European Space enci's efforts to fight air pollution using a fleet of Copernicus satellites, 84 00:08:05.800 --> 00:08:11.839 and later in the science report any study shows that open ais chat GPT 85 00:08:11.160 --> 00:08:16.399 four can answer medical questions better than most doctors. All that and more still 86 00:08:16.439 --> 00:08:37.559 to come on spacetime. Air pollution has become a serious global problem. According 87 00:08:37.600 --> 00:08:43.039 to the World Health Organization, seven million premature deaths each year are linked to 88 00:08:43.080 --> 00:08:50.320 air pollution. The European space edencies Copernicus atmosphere monitoring services provide key information about 89 00:08:50.360 --> 00:08:56.279 air quality around the globe, helping sitists understand how people can keep the planet 90 00:08:56.360 --> 00:09:03.000 breathable. Using the European Space ensies constantly of Earth observation satellites, it provides 91 00:09:03.080 --> 00:09:07.320 constant, quality controlled information in relation to air pollution and health, solar energy, 92 00:09:07.480 --> 00:09:13.559 greenhouse gases, and climate across the globe. Over the past year alone, 93 00:09:13.679 --> 00:09:18.200 East's Sentinel satellite constellation has been monitoring Sahara desert dust storms, crossing the 94 00:09:18.240 --> 00:09:24.519 Iberian Peninsula and then continuing across France towards northern and Central Europe. The networks 95 00:09:24.559 --> 00:09:31.080 also observed large scale smoke blooms generated by devastating wildfires in central and southern Chilee 96 00:09:31.200 --> 00:09:35.759 that was back in February, which then spread over the Pacific Ocean. They 97 00:09:35.799 --> 00:09:41.480 followed even more extreme wildfires back in December, which burnt across Europe and the 98 00:09:41.519 --> 00:09:48.559 Americas. The sentinel's atmospheric monitoring systems tracked devastating wildfire emissions around the globe, 99 00:09:48.840 --> 00:09:54.759 and it monitored as European wildfire emissions reached their highest levels in fifteen years. 100 00:09:54.399 --> 00:10:00.919 Copernicus also studied persistent Atarctic ozone holes in twenty twenty, twenty twenty one, 101 00:10:01.080 --> 00:10:05.440 and twenty twenty two, which included some unusual behavior. Not only did the 102 00:10:05.480 --> 00:10:09.240 closure of the ozone whole last year take longer than usual, but it was 103 00:10:09.320 --> 00:10:15.559 also relatively large. The constellation also kept a close siye on the evolution of 104 00:10:15.600 --> 00:10:20.080 Monoloa's volcanic eruption in Hawaii, which wound up spewing out a huge plume of 105 00:10:20.200 --> 00:10:26.519 soph of the oxide, and it issued early warnings about spring dust, air 106 00:10:26.519 --> 00:10:31.600 pollution and pollen outbreaks across Central Europe, and thanks to Copernicus, scientists are 107 00:10:31.600 --> 00:10:37.200 now predicting further very high levels of surface ozone pollution across large areas of Europe. 108 00:10:37.279 --> 00:10:45.519 It's temperature starts saw again this report from a TV. The atmosphere crucial 109 00:10:45.559 --> 00:10:50.639 for life on Earth. It makes our planet habitable. However, air pollution 110 00:10:50.679 --> 00:10:56.360 has become a serious problem. Human activities are altering our atmosphere at an unprecedented 111 00:10:56.399 --> 00:11:01.240 scale. This has an impact on air quality is affecting people's health all around 112 00:11:01.279 --> 00:11:07.200 the globe. Studies even show that in the urbanized regions of Europe, ninety 113 00:11:07.279 --> 00:11:11.440 percent of the population is exposed to harmful levels of air pollution. According to 114 00:11:11.720 --> 00:11:18.480 the World Earth Organization, adverse equality is responsible of seven million premature of death 115 00:11:18.600 --> 00:11:24.120 on a global scale. In Europe, the Environment Agency is estimate to four 116 00:11:24.279 --> 00:11:31.120 hundred to five hundred thousand premature death occurring. So this is a very real 117 00:11:31.480 --> 00:11:35.720 problem. Air Quality affects our health in many ways, from causing heart and 118 00:11:35.840 --> 00:11:43.080 respiratory disease, to exacerbating asthma and even impacting cognitive abilities. Recent studies have 119 00:11:43.159 --> 00:11:48.679 also shown that even relatively low levels of air pollution can affect people's health. 120 00:11:48.519 --> 00:11:54.360 In order to better understand this problem and find solutions, scientists need reliable data 121 00:11:54.480 --> 00:11:58.480 on where and how the air is being polluted and how it spreads. This 122 00:11:58.639 --> 00:12:03.879 can be done locally by taking in situ measurements, but with satellite data this 123 00:12:05.039 --> 00:12:09.600 can be achieved at a larger scale across regions and countries, thus tracking pollution 124 00:12:09.879 --> 00:12:15.080 as it's spread by the wind. Over the last two decades, the European 125 00:12:15.159 --> 00:12:20.240 Space Agency launched a number of satellites the to observe our atmosphere the measure air 126 00:12:20.320 --> 00:12:26.759 pollutants. ESA is involved in the air quality measurements by providing space measurements satellites 127 00:12:26.840 --> 00:12:31.399 that have the capability to measure air pollution trace causes. We started on this 128 00:12:31.799 --> 00:12:35.639 during ninety five by putting Come in space this instrumentation. This capability has then 129 00:12:35.720 --> 00:12:41.919 been improved step by step by follow on instruments like Eskimaki two or OMI and 130 00:12:41.039 --> 00:12:46.799 now with the Copernicus mission Sentinel five precursor, we have clover coverage within one 131 00:12:46.919 --> 00:12:52.360 day and this allows for the first time to use this space based measurements to 132 00:12:52.480 --> 00:12:56.960 be implemented in operational services for air quality monitoring but also for forecasting. The 133 00:12:58.080 --> 00:13:01.840 Copernicus program, with its Sentinels satellites, has bridged the gap between research and 134 00:13:01.960 --> 00:13:09.600 operational services. Using both satellite and institution measurements, the Copernica's atmosphere monitoring services 135 00:13:11.159 --> 00:13:16.679 provides information about air composition and quality across the globe at the European level. 136 00:13:16.960 --> 00:13:22.840 It's almost like a weather forecast. This information is picked up by media and 137 00:13:22.000 --> 00:13:28.799 smartphone applications, warning the public on poor air quality in their region. With 138 00:13:28.000 --> 00:13:33.960 Copernica's Sentinel five P launched in twenty seventeen, which carries a novel instrument called 139 00:13:35.000 --> 00:13:39.480 Tropome, a big step was taken to observe sources of air pollution. Now 140 00:13:39.559 --> 00:13:46.799 pollution hotspots are clearly visible. This increased spatial resolution comes from Sentinel five P 141 00:13:46.000 --> 00:13:52.879 spectrometer the Tropom instrument. The Tropome instrument is really a big step forward in 142 00:13:52.000 --> 00:13:56.279 what we hear, what we can do, and that's for different reasons. 143 00:13:56.440 --> 00:14:01.399 I think maybe the most striking when you look at the data is the very 144 00:14:01.519 --> 00:14:07.320 good space or resolution. So it can measure with a ground resolution of about 145 00:14:07.639 --> 00:14:11.080 three and a half by seven kilometers, which is more an effector of ten 146 00:14:11.240 --> 00:14:18.519 better than what we can that previous instruments had. With its Tropome instrument, 147 00:14:18.039 --> 00:14:24.000 Copernicus Sentinel five P is showing how important it is by delivering operational data on 148 00:14:24.120 --> 00:14:31.919 air quality every day within the Kaupernicus program, ESA and the EU are already 149 00:14:31.000 --> 00:14:37.039 preparing new missions to monitor our atmosphere. For example, one of the potential 150 00:14:37.159 --> 00:14:43.960 future Sentinel missions has been proposed to monitor emissions of carbon dioxide. Our atmosphere 151 00:14:43.120 --> 00:14:48.240 is precious, but polluting emissions are both responsible for poor air quality as well 152 00:14:48.320 --> 00:14:56.320 as adding to the greenhouse effect driving climate change. By monitoring emissions and understanding 153 00:14:56.360 --> 00:15:01.000 how they spread, we might find solutions to keep our planet breathable unsustainable. 154 00:15:01.120 --> 00:15:05.240 And in that report from ESA TV, we heard from Vincent Unripew ahead of 155 00:15:05.320 --> 00:15:11.440 Copernicus Atmospheric Monitoring Services, Sentinel five piece mission manager at ESA, Klaus Zena, 156 00:15:11.639 --> 00:15:16.720 and Peepin Vifkin, one of the principal investigators with the Sentinel five P 157 00:15:16.960 --> 00:15:37.879 mission. This space time and time out to take another brief look at some 158 00:15:37.919 --> 00:15:41.559 of the other stories making news in science this week. With a Science report, 159 00:15:41.559 --> 00:15:46.000 A new study is found that sticking to a specific type of Mediterranean diet 160 00:15:46.200 --> 00:15:52.679 known as a mind diet may help reduce dementia risk. Mind diet stands for 161 00:15:52.840 --> 00:16:00.320 Mediterranean dietary approaches to stop hypertension intervention for neurodegenerative delay. The findings were ordered 162 00:16:00.320 --> 00:16:03.440 in the Journal of the American Medical Association. Looked at two hundred thousand people, 163 00:16:03.720 --> 00:16:08.279 finding that those who stuck to the diet had a seventeen percent lower risk 164 00:16:08.320 --> 00:16:15.200 of dementia. The study showed that this type of diet emphasizes natural protein based 165 00:16:15.279 --> 00:16:19.200 foods, limits the intake of animal foods and foods high and saturated fats, 166 00:16:19.559 --> 00:16:26.200 and uniquely encourages people to eat berries and green leafy vegetables rich in vitaments and 167 00:16:26.279 --> 00:16:32.960 antioxidants. Paleontologists have found that, just like modern day elephants, male wooly 168 00:16:33.080 --> 00:16:37.960 mammoths underwent musks, which are testosterone driven changes during the mating season. A 169 00:16:38.159 --> 00:16:44.639 report of the journal Nature claims scientists detected hormone fluctuations in the dentine of male 170 00:16:44.759 --> 00:16:48.200 mammoth tasks estimated to be roughly thirty nine to thirty three thousand years old, 171 00:16:48.559 --> 00:16:52.840 and there were similar to hormonal patterns seen in the tusks of modern day male 172 00:16:52.960 --> 00:16:59.799 African elephants and whatever. Scientists say the hormone spikes were smaller in the mammoths 173 00:17:00.320 --> 00:17:04.640 because the mammoth tusks they tested had degraded over thousands of years. But they 174 00:17:04.680 --> 00:17:11.000 also tested female mammoth tusks aged around six thousand years old and found no fluctuation 175 00:17:11.079 --> 00:17:15.519 in testosterone levels. The findings of the first evidence of must in mammoths based 176 00:17:15.559 --> 00:17:21.440 on hormonal data, demonstrating the potential of teeth and tusks for shitting new light 177 00:17:21.640 --> 00:17:26.599 on the live history of events in extinct animals. A new study has shown 178 00:17:26.680 --> 00:17:33.640 that chat gpt is capable of answering questions about medical advice at a standard that's 179 00:17:33.680 --> 00:17:37.200 often higher than what a doctor can provide, assuming, of course, the 180 00:17:37.359 --> 00:17:41.799 AIS being honest. The findings are reported in the Journal of the American Medical 181 00:17:41.839 --> 00:17:47.480 Association, suggest that AIS could play a role in helping overwarm doctors draft better, 182 00:17:47.640 --> 00:17:52.240 quicker responses to questions. The authors simply asked the chat gpt to answer 183 00:17:52.359 --> 00:17:57.799 questions submitted to Reddit's ask stocks that have already been answered by verified doctors, 184 00:17:59.480 --> 00:18:03.839 and the team of licensed health professionals then judged the response to determine which had 185 00:18:03.880 --> 00:18:08.960 the better quality of information and better bedside manner. The authors found that the 186 00:18:10.119 --> 00:18:14.799 chat GPT's response turned out being better for seventy eight point six percent of the 187 00:18:14.880 --> 00:18:18.640 questions answered Now, of course it read it threads, not the same as 188 00:18:18.680 --> 00:18:22.880 a doctor's office, but the results do show that AIS could help doctors deliver 189 00:18:23.039 --> 00:18:30.319 information in the best easiest way possible in the future. Now, while we're 190 00:18:30.359 --> 00:18:34.400 on the subject of artificial intelligence, Jeffrey Hinton, the man widely seen as 191 00:18:34.440 --> 00:18:40.079 the father of artificial intelligence, has Quitty's job with Google, warning about the 192 00:18:40.160 --> 00:18:45.039 growing dangers from developments in AI. The seventy five year old announced his resignation 193 00:18:45.160 --> 00:18:49.359 in the statement to The New York Times, saying he now regretted his work, 194 00:18:51.000 --> 00:18:56.359 and later telling the BBC that the dangers of AI chatbots are quite scary. 195 00:18:56.079 --> 00:19:03.039 Hinton's pioneering research on deep learning in neural networks pave the way for today's 196 00:19:03.079 --> 00:19:07.960 AI systems like chat GPT four. Hitton's major AI breakthrough came back in twenty 197 00:19:08.079 --> 00:19:14.519 twelve when his team successfully created an algorithm which could analyze images in the process 198 00:19:14.720 --> 00:19:18.839 identifying common elements such as people, dogs, or cars. Now, that 199 00:19:19.000 --> 00:19:23.599 was only a rudimentary algorithm, but it would eventually lead to today's current level 200 00:19:23.680 --> 00:19:30.599 of AIS like open AI, chat, GPT four and Google's bad Hitten warns 201 00:19:30.680 --> 00:19:34.759 that AIS will soon be more intelligent than humans if they aren't already and they're 202 00:19:34.759 --> 00:19:41.119 already eclipsing humans in the amount of general knowledge they hold. Hitten's fees also 203 00:19:41.160 --> 00:19:45.359 worker those expressed in a public letter earlier this year by more than a thousand 204 00:19:45.519 --> 00:19:49.640 other tech leaders, including Elon Musk and Steve Wasn't the act. They called 205 00:19:49.720 --> 00:19:55.279 for a six month halt in AI development because of concerns over its use. 206 00:19:56.079 --> 00:20:00.839 Hitten says he's especially worried about bad actors like Russian president that to me Putin, 207 00:20:00.000 --> 00:20:03.680 who could use AI for bad things, such as getting more power. 208 00:20:04.400 --> 00:20:11.720 Technology editor alexaharav Royt from Take Advice, Start Life says his actions are significance. 209 00:20:11.880 --> 00:20:15.319 It's interesting that someone says that they regret their life's work. I mean, 210 00:20:15.400 --> 00:20:19.160 that's a pretty amazing statement. He did get his PhD and artificial intelligence 211 00:20:19.279 --> 00:20:22.440 forty five years ago, so human kind has been thinking about this for a 212 00:20:22.559 --> 00:20:26.279 long long time. Originally just something that was science fiction, but here nearly 213 00:20:26.319 --> 00:20:30.480 fifty years later, and it's becoming true. And he did say that he 214 00:20:30.559 --> 00:20:33.720 consoles himself with a normal excuse. If he hadn't done it, he says, 215 00:20:33.720 --> 00:20:36.039 somebody else would have it. It's hard to see how he can prevent 216 00:20:36.119 --> 00:20:38.440 bad actors from using it for bad things. What's he scared of? Is 217 00:20:38.480 --> 00:20:41.319 it the Skynet thing? Look, I think that's a worried go on the 218 00:20:41.400 --> 00:20:44.920 track. But at the moment he fears that the Internet could be flooded with 219 00:20:45.079 --> 00:20:48.240 misinformation. Come on, that's the ABC and CNN's job. Sure well, 220 00:20:48.279 --> 00:20:52.200 we haven't had fake news for quite some time without the power of AI. 221 00:20:52.279 --> 00:20:56.480 We've had that through organic intelligence. But Hinton was concerned that we just won't 222 00:20:56.519 --> 00:21:00.319 know what's true. We have had the case just in the last couple of 223 00:21:00.400 --> 00:21:04.200 weeks where a German magazine called The Actuel published a world first interview with Michael 224 00:21:04.240 --> 00:21:08.359 Schumacher and hinted that it was deceptively real. And it turns out when people 225 00:21:08.400 --> 00:21:12.880 read the actual interview that the whole thing was generated by something from character dot 226 00:21:14.000 --> 00:21:17.640 AI, which allows you to talk to famous people who are either no longer 227 00:21:17.680 --> 00:21:21.160 with us or, in Michael Schumacher's case, are obviously not able to speak 228 00:21:21.200 --> 00:21:23.119 anymore. He said in the New York Times article that everyone's been quoting. 229 00:21:23.160 --> 00:21:26.279 He says, the idea that this stuff could actually gets smarter than people. 230 00:21:26.400 --> 00:21:30.119 A few people believed that, he says, but most people thought it was 231 00:21:30.240 --> 00:21:32.440 way off, and I thought it was way off. I thought it was 232 00:21:32.559 --> 00:21:36.079 thirty to fifty years or even longer away. Obviously, he says, I 233 00:21:36.279 --> 00:21:40.440 no longer think that. And in fact, the article seemed to imply that 234 00:21:40.759 --> 00:21:42.680 Hinton left Google so that he could criticize Google. But he said in the 235 00:21:42.720 --> 00:21:45.799 tweet that he left so that he could talk about the dangers of AI without 236 00:21:45.880 --> 00:21:49.960 considering how this impacts Google. He reckons at Google has acted very responsibly. 237 00:21:51.119 --> 00:21:55.039 Now. The thing is that AI has already gone mad in certain circumstances. 238 00:21:55.119 --> 00:21:59.519 Microsoft has had to put the brakes on its own version of chat GBT for 239 00:21:59.640 --> 00:22:03.680 the Big AI. It's also become woken politically biased. That's all down other 240 00:22:03.720 --> 00:22:07.759 people who are of course programming. Yeah, absolutely absolutely. I mean this 241 00:22:07.920 --> 00:22:11.880 is where individuals and companies are going to have to end up with their own 242 00:22:11.200 --> 00:22:15.319 systems, their own AI systems that can be filled with the information that you 243 00:22:15.400 --> 00:22:18.799 have read it. I mean, already Samsung has banned its staff from using 244 00:22:18.880 --> 00:22:23.640 chat GPT. Samsung had discovered that some of its staff had been putting sensitive 245 00:22:23.720 --> 00:22:29.960 information about unreleased products and also into chat tiput and also asking it to provide 246 00:22:30.000 --> 00:22:33.960 summaries of meetings where sensitive topics were discussed, and this information is being used 247 00:22:34.000 --> 00:22:37.920 by chat GPT without chat tiputs knowledge that Samsung will necessarily put it in there, 248 00:22:37.960 --> 00:22:41.559 but it's just being used to train the system and could easily be regurgitated 249 00:22:41.680 --> 00:22:45.480 in answers that it gives to other people, because that's how these chatbots work. 250 00:22:45.559 --> 00:22:52.160 They are incredibly proficient aggregation and regurgitation machines. They give the appearance of 251 00:22:52.240 --> 00:22:56.000 being intelligent because they've got so much information at their disposal that they know what, 252 00:22:56.319 --> 00:23:00.440 in theory the correct things are to say in regards saying response, But 253 00:23:00.480 --> 00:23:03.359 again it all depends what has been input into the system. And another concern 254 00:23:03.640 --> 00:23:07.759 is that now we've got AI systems which can even read your mind. Yeah, 255 00:23:07.880 --> 00:23:14.519 they're getting individuals to train themselves on a system that can analyze the patterns 256 00:23:14.559 --> 00:23:18.240 of their brain and then they can take this analysis run it through a GPT. 257 00:23:18.440 --> 00:23:21.640 It's a version one GPT, not the three point four or five years 258 00:23:21.680 --> 00:23:26.720 today, and the GPTs results or responses Hugh very closely to what it is 259 00:23:26.799 --> 00:23:30.559 that they say they were thinking about all what they were seeing in a video. 260 00:23:30.759 --> 00:23:33.880 So this is the beginning of actually being able to have technology read your 261 00:23:33.920 --> 00:23:38.839 mind and give out results. The great minds that Apple and Google have come 262 00:23:38.920 --> 00:23:45.240 together and they've discovered whoop you do? People don't like being tracked? Why 263 00:23:45.279 --> 00:23:48.920 did it attack them so long? It's not that people didn't like tracking. 264 00:23:48.000 --> 00:23:52.119 I mean you were able to track your friends on iPhones. Using to Find 265 00:23:52.200 --> 00:23:56.960 my Friends, you could switch on the ability to share your location for an 266 00:23:56.000 --> 00:23:59.920 hour or for the rest of the day, or even on an ongoing basis. 267 00:24:00.160 --> 00:24:03.519 Arents use these tracking features. Let me ask you a question here, 268 00:24:03.599 --> 00:24:06.160 do you really if your tracker on or do you turn it off like I 269 00:24:06.279 --> 00:24:08.200 turn off? I turn it off. But if I'm going to a friends 270 00:24:08.240 --> 00:24:11.920 place, for example, and then I might turn it on for the hour 271 00:24:11.000 --> 00:24:15.359 so they can see how long I'm going to be, that can actually tell 272 00:24:15.559 --> 00:24:18.200 the distance I am away. I use this with my sister the other day. 273 00:24:18.200 --> 00:24:19.839 She was coming to visit with her boyfriend, and I told her, 274 00:24:19.880 --> 00:24:22.279 you know, turn your tracking on for the next hour, because they were 275 00:24:22.279 --> 00:24:26.039 about narrow way and it was really cool to be able to see ow they're 276 00:24:26.039 --> 00:24:27.279 ten minutes away, they're outside, and you know, you could sort of 277 00:24:27.400 --> 00:24:30.400 not have to worry about where they were. Now. The problem is when 278 00:24:30.599 --> 00:24:34.559 people have been using the Tile and the Apple trackers to track people, to 279 00:24:34.680 --> 00:24:38.640 track ex partners, to track cars, to store people. And Apple has 280 00:24:38.680 --> 00:24:42.200 been the only company that actually alerts you to the fact that you have a 281 00:24:42.279 --> 00:24:45.720 tracker that is not one of your trackers, that seems to be staying with 282 00:24:45.799 --> 00:24:48.440 you, is in your area. The Tile tracker doesn't do that. Other 283 00:24:48.480 --> 00:24:52.640 trackers haven't done that, and so Apple has actually been responsible in this area 284 00:24:52.799 --> 00:25:00.119 by alerting people though even delivered an app that Android users can install to see 285 00:25:00.160 --> 00:25:03.079 if there's any Apple trackers around them. But Apple and Google have jointless submitted 286 00:25:03.119 --> 00:25:08.039 a proposed industry specification. This is to help combat the misuse of Bluetooth location 287 00:25:08.119 --> 00:25:12.000 tracking devices to unwanted tracking. And that's the first of us kind specification. 288 00:25:12.240 --> 00:25:33.240 That's Alexahara Royd from Tech Advice, Start Live, and that's the show for 289 00:25:33.400 --> 00:25:37.599 now. Spacetime is available every Monday, Wednesday and Friday through Apple Podcasts, 290 00:25:37.680 --> 00:25:45.079 iTunes, Stitcher, Google podcast pocket Casts, Spotify a Cast, Amazon Music, 291 00:25:45.319 --> 00:25:49.559 Bytes dot Com, SoundCloud YouTube, your favorite podcast download provider, and 292 00:25:49.799 --> 00:25:56.400 from Spacetime with Stewart Gary dot Com. Spacetime's also broadcast through the National Science 293 00:25:56.480 --> 00:26:00.960 Foundation on Science Zone Radio and on both iHeartRadio and tune In Radio. And 294 00:26:02.119 --> 00:26:06.079 you can help to support our show by visiting the Spacetime Store for a range 295 00:26:06.119 --> 00:26:11.119 of promotional merchandising goodies, or by becoming a Spacetime Patron, which gives you 296 00:26:11.200 --> 00:26:15.160 access to triple episode commercial free versions of the show, as well as lots 297 00:26:15.160 --> 00:26:18.599 of burnus audio content which doesn't go to wear, access to our exclusive Facebook 298 00:26:18.640 --> 00:26:23.079 group, and other rewards. Just go to Spacetime with Stewart Gary dot com 299 00:26:23.319 --> 00:26:27.160 for full details, and if you want more Spacetime, please check out our 300 00:26:27.200 --> 00:26:30.279 blog where you'll find all the stuff we couldn't fit in the show, as 301 00:26:30.359 --> 00:26:34.400 well as heaps of images, news stories, loads of videos and things on 302 00:26:34.519 --> 00:26:38.960 the whereby find interesting or amusing. Just go to Spacetime with Stewart Gary dot 303 00:26:40.039 --> 00:26:44.640 Tumbler dot com. That's all one word, and that's Tumbler without the E. 304 00:26:45.359 --> 00:26:48.279 You can also follow us through at Stewart Garry on Twitter, at Spacetime 305 00:26:48.319 --> 00:26:52.799 with Stewart Garry, on Instagram, through our Spacetime YouTube channel, and on 306 00:26:52.960 --> 00:26:57.319 Facebook. Just go to Facebook dot com, forward slash, Spacetime with Stewart 307 00:26:57.359 --> 00:27:03.079 Gary and Spacetime. It's brought to you in collaboration with Australian Sky and Telescope 308 00:27:03.119 --> 00:27:07.759 magazine. You're a window on the universe. You've been listening to Spacetime with 309 00:27:07.920 --> 00:27:11.960 Stuart Gary. This has been another quality podcast production from bytes dot com.