1 00:00:00,080 --> 00:00:02,919 Speaker 1: A decade ago, one of the most technologically advanced commercial 2 00:00:02,919 --> 00:00:05,759 aircraft in the world, a Boeing seven to seven sevens 3 00:00:05,759 --> 00:00:08,880 Hearing two hundred and thirty nine souls, executed a series 4 00:00:08,880 --> 00:00:11,439 of maneuvers that took it violently off course and into 5 00:00:11,439 --> 00:00:15,320 the void. It vanished from civilian radar shortly after takeoff 6 00:00:15,359 --> 00:00:18,719 from Qualumpor, and ever since MH three to seventy has 7 00:00:18,760 --> 00:00:22,079 stood as the greatest aviation mystery of our time. It's 8 00:00:22,120 --> 00:00:25,800 a puzzle that has defied two massive international search efforts. 9 00:00:26,039 --> 00:00:27,640 Speaker 2: It really has. It is the ghost plane of the 10 00:00:27,679 --> 00:00:32,759 Indian Ocean, a symbol of information asymmetry. We know where 11 00:00:32,759 --> 00:00:34,719 it wasn't, but we still don't know where it is. 12 00:00:34,600 --> 00:00:36,759 Speaker 1: And we definitely don't know why, and we certainly. 13 00:00:36,439 --> 00:00:38,920 Speaker 2: Don't know why. The previous search missions, I mean, despite 14 00:00:38,960 --> 00:00:42,119 their scale, they just met with profound frustration and failure, 15 00:00:42,320 --> 00:00:46,280 leaving us with what just fragmented debris in these whispers 16 00:00:46,320 --> 00:00:47,679 from satellite data. 17 00:00:47,719 --> 00:00:50,079 Speaker 1: But hold on to that sense of enduring mystery, because 18 00:00:50,079 --> 00:00:52,320 today we are delving into what is now the third 19 00:00:52,359 --> 00:00:55,359 major and potentially final search attempt for MH three to seventy. 20 00:00:56,359 --> 00:00:59,880 Welcome to thrilling threads where we take the crucial information 21 00:01:00,159 --> 00:01:02,079 and spin it into actionable knowledge for you. 22 00:01:02,520 --> 00:01:06,200 Speaker 2: Our mission today is to really dissect this high stakes, 23 00:01:06,519 --> 00:01:10,200 high tech expedition that's currently underway. We are grounding this 24 00:01:10,280 --> 00:01:13,319 discussion in a recent CNN report. It was called new 25 00:01:13,439 --> 00:01:17,120 MH three seventy Search underway and it features some really 26 00:01:17,159 --> 00:01:21,040 critical input from aviation safety analyst David Suzi and the 27 00:01:21,519 --> 00:01:23,200 legendary oceanographer David Gallop. 28 00:01:23,359 --> 00:01:27,079 Speaker 1: Yeah, and our goal is to analyze the new arsenal 29 00:01:27,120 --> 00:01:31,640 being deployed. We're talking artificial intelligence, cutting edge swarm, robotics, 30 00:01:31,680 --> 00:01:35,680 and this rigorous re analysis of revised flight data. We 31 00:01:35,719 --> 00:01:38,799 want to understand what makes this attempt fundamentally different from 32 00:01:38,840 --> 00:01:41,879 the previous two failures. This isn't just a retry, it's 33 00:01:41,879 --> 00:01:44,560 a completely new technical approach to the problem. 34 00:01:44,599 --> 00:01:47,120 Speaker 2: And here's the core contradiction we have to explore because 35 00:01:47,159 --> 00:01:50,239 it provides the essential pension for this whole discussion. We 36 00:01:50,280 --> 00:01:54,040 have one expert, David Susi, who went from being profoundly 37 00:01:54,040 --> 00:01:57,280 skeptical to being very confident that they will find the wreckage. 38 00:01:57,359 --> 00:01:58,560 Speaker 1: He thinks they've solved it. 39 00:01:58,920 --> 00:02:01,359 Speaker 2: He believes the new tool have solved the data problem. 40 00:02:01,599 --> 00:02:06,079 Speaker 1: Meanwhile, you have the vastly experienced ocean explorer David Gallo, 41 00:02:06,280 --> 00:02:10,560 who is well, he's far more circumspect. He stresses that 42 00:02:10,639 --> 00:02:14,000 the key challenge isn't the quality of the instruments, it's 43 00:02:14,080 --> 00:02:16,599 the brute reality of determining if they are looking in 44 00:02:16,639 --> 00:02:18,120 the right haystack in the first place. 45 00:02:18,199 --> 00:02:19,479 Speaker 2: And that's the real tension. 46 00:02:19,560 --> 00:02:23,280 Speaker 1: Is that tension between technological optimism and the brutal reality 47 00:02:23,280 --> 00:02:27,039 of deep sea exploration is precisely what we are unpacking 48 00:02:27,039 --> 00:02:29,719 for you today. Let's get into it, Okay, let's unpack 49 00:02:29,800 --> 00:02:34,120 this shift in strategy and really in confidence. We are 50 00:02:34,159 --> 00:02:37,199 starting with the renewed hunt, which is being spearheaded by 51 00:02:37,199 --> 00:02:41,800 the Texas based marine robotics firm ocean Infinity. This marks 52 00:02:41,879 --> 00:02:45,199 their third time taking on this monumental task of finding 53 00:02:45,280 --> 00:02:48,479 the long lost plane. But and this is key, they 54 00:02:48,479 --> 00:02:51,240 are approaching it with completely new hardware and software. 55 00:02:51,439 --> 00:02:54,560 Speaker 2: And ocean Infinity is not just another salvage operator. I mean, 56 00:02:54,560 --> 00:02:57,000 this is a firm that has staked his reputation on 57 00:02:57,120 --> 00:03:00,439 leveraging cutting edge autonomous technology. The fact that they're committing 58 00:03:00,439 --> 00:03:03,400 to a third attempt, and possibly under a no fine, 59 00:03:03,439 --> 00:03:06,960 no fee arrangement that speaks volumes about their confidence in 60 00:03:07,000 --> 00:03:10,400 their new tech stack and the revised data models. They are, 61 00:03:10,560 --> 00:03:14,000 without a doubt, the market leader in these large scale 62 00:03:14,039 --> 00:03:15,319 deep sea surveys now. 63 00:03:15,439 --> 00:03:19,000 Speaker 1: And what's truly fascinating to me is the transformation and 64 00:03:19,120 --> 00:03:22,479 perspective from someone who has followed this mystery for a decade, 65 00:03:22,599 --> 00:03:25,039 like the aviation safety analyst David Sussi. 66 00:03:25,199 --> 00:03:26,840 Speaker 2: Yeah, his change of heart is really something. 67 00:03:26,960 --> 00:03:30,400 Speaker 1: He was quoted as being really apprehensive about these continued 68 00:03:30,439 --> 00:03:33,599 searches until now. And this wasn't just some casual observation. 69 00:03:34,000 --> 00:03:36,719 Susie co authored a book about MH three seventy back 70 00:03:36,719 --> 00:03:40,319 in twenty fifteen, so he knows every single failure point, 71 00:03:40,560 --> 00:03:44,719 every false lead. Intimately, we need to understand the mechanism 72 00:03:44,759 --> 00:03:48,080 of his conversion. Why the sudden surge and confidence. 73 00:03:48,240 --> 00:03:51,599 Speaker 2: Well, he cited three very specific integrated factors that have 74 00:03:51,639 --> 00:03:54,120 converged to give them the tactical advantage they just didn't 75 00:03:54,159 --> 00:03:57,240 have before. He sees this attempt not as a third 76 00:03:57,280 --> 00:04:00,639 spin of the wheel, but as the first search using 77 00:04:00,680 --> 00:04:02,800 adequate twenty first century tools. 78 00:04:02,520 --> 00:04:04,520 Speaker 1: And applied to a smaller target zone. 79 00:04:04,520 --> 00:04:07,439 Speaker 2: Exactly applied to a newly defined, much smaller target zone. 80 00:04:07,599 --> 00:04:10,560 Speaker 1: De tail those three factors for us, because they are 81 00:04:10,639 --> 00:04:13,800 the engine driving the current optimism and the reason Ocean 82 00:04:13,800 --> 00:04:16,360 Affinity is deploying millions of dollars in resources. 83 00:04:16,759 --> 00:04:19,879 Speaker 2: Okay, so the first factor is the revised data analysis, 84 00:04:20,040 --> 00:04:24,000 which has been well, it's been supercharged by artificial intelligence. 85 00:04:24,480 --> 00:04:27,680 Susie noted explicitly that the ability to use AI for 86 00:04:27,800 --> 00:04:32,120 high volume data analysis simply was not available for previous searches, 87 00:04:32,279 --> 00:04:35,480 especially not the first one, especially the one conducted in 88 00:04:35,519 --> 00:04:38,720 the immediate aftermath of the disappearance. It just wasn't there. 89 00:04:38,879 --> 00:04:41,040 Speaker 1: But let me probe that for a moment. If the 90 00:04:41,079 --> 00:04:44,920 previous searches collected raw data that was you know, maybe 91 00:04:44,959 --> 00:04:49,279 flood noisy or misinterpreted by human operators, how does AI 92 00:04:49,480 --> 00:04:53,079 suddenly make that data useful? Is the AI just confirming 93 00:04:53,120 --> 00:04:54,240 the mistakes of the past. 94 00:04:54,399 --> 00:04:56,680 Speaker 2: That is the critical distinction, and it's a great question. 95 00:04:57,240 --> 00:04:59,600 The AI isn't finding the plane on its own, it's 96 00:04:59,639 --> 00:05:03,639 finding the anomalies. The sheer volume of sonar returns that 97 00:05:03,759 --> 00:05:06,120 a lot of data. So, as you mentioned, from the 98 00:05:06,160 --> 00:05:09,279 previous searches, it just bordered on information overload. I can 99 00:05:09,360 --> 00:05:13,199 imagine you had human operators, you know, despite their expertise. 100 00:05:13,480 --> 00:05:16,600 We had to filter out billions of sound pixels related 101 00:05:16,600 --> 00:05:20,439 to geology, to deep sea currents, marine life. Right AI 102 00:05:20,600 --> 00:05:24,720 can be trained on known signatures of aviation debris, and crucially, 103 00:05:25,040 --> 00:05:27,920 it can look at every historical data point without fatigue 104 00:05:28,120 --> 00:05:31,279 and without the assumption that an area has already been cleared. 105 00:05:31,480 --> 00:05:35,040 Speaker 1: So it's not relying on the past human interpretation of 106 00:05:35,079 --> 00:05:38,720 the data exactly, it's rerunning the raw noisy data through 107 00:05:38,720 --> 00:05:43,000 a powerful objective filter. This helps minimize the risk that 108 00:05:43,040 --> 00:05:45,480 the recage was actually seen back in twenty fourteen or 109 00:05:45,480 --> 00:05:48,759 twenty seventeen, but was just miscategorized as a rock formation 110 00:05:48,959 --> 00:05:49,399 or something. 111 00:05:49,480 --> 00:05:52,399 Speaker 2: That's it. It's quality control on a massive scale. When 112 00:05:52,439 --> 00:05:54,480 you are mapping an area the size of a state, 113 00:05:54,920 --> 00:05:58,560 the raw acoustic data is just astronomical. The AI's ability 114 00:05:58,600 --> 00:06:03,439 to flag specific characteristics angular edges, middle density variations that 115 00:06:03,560 --> 00:06:06,040 might have been dismissed as noise by a tired operator. 116 00:06:06,439 --> 00:06:08,839 That's what gives this new attempt its analytical struck. 117 00:06:08,920 --> 00:06:10,839 Speaker 1: Okay, that makes a lot of sense. So what's the 118 00:06:10,879 --> 00:06:13,040 second factor driving Susi's confidence. 119 00:06:13,160 --> 00:06:16,639 Speaker 2: The technology of the hunt itself new swarm technology. 120 00:06:17,319 --> 00:06:20,680 Speaker 1: Instead of relying on that extremely slow serial method of 121 00:06:20,680 --> 00:06:23,519 towing a single device from a ship. They are now 122 00:06:23,560 --> 00:06:28,639 deploying multiple autonomous underwater vehicles AUVs at the same time, 123 00:06:29,120 --> 00:06:31,360 coordinated across a wide search grid. 124 00:06:32,199 --> 00:06:34,040 Speaker 2: We are definitely going to dedicate a whole section to 125 00:06:34,079 --> 00:06:36,879 the mechanics of the swarm because it's fascinating. But just 126 00:06:37,199 --> 00:06:42,160 knowing that they are multiplying their search capability exponentially, that 127 00:06:42,279 --> 00:06:45,720 fundamentally changes the economics and the time constraints of the search. 128 00:06:45,879 --> 00:06:49,000 Speaker 1: It absolutely does. It moves the effort from serial searching 129 00:06:49,000 --> 00:06:52,000 to parallel searching, which gives them a breadth of coverage 130 00:06:52,040 --> 00:06:53,240 they never ever had before. 131 00:06:53,399 --> 00:06:54,360 Speaker 2: And the third factor. 132 00:06:54,439 --> 00:06:57,120 Speaker 1: The third factor addresses the fundamental problem of where the 133 00:06:57,160 --> 00:07:00,279 plane actually went. Improved ocean current mapping. 134 00:07:00,199 --> 00:07:03,040 Speaker 2: And this involves leveraging the physical evidence that was actually 135 00:07:03,079 --> 00:07:04,439 found correct precisely. 136 00:07:04,800 --> 00:07:08,040 Speaker 1: They've had the benefit of analyzing actual parts of the wreckage, 137 00:07:08,120 --> 00:07:11,439 the debris that washed up, most notably that flapparon found 138 00:07:11,519 --> 00:07:14,759 on Reuenion Island and other fragments recovered near Madagascar. 139 00:07:15,000 --> 00:07:19,040 Speaker 2: They've used new technology to remap the historical three dimensional 140 00:07:19,079 --> 00:07:22,759 ocean current flows with far greater precision than they could 141 00:07:22,800 --> 00:07:23,600 have ten years ago. 142 00:07:24,000 --> 00:07:27,879 Speaker 1: That is a fantastic example of using forensic evidence to 143 00:07:27,959 --> 00:07:31,879 constrain the mathematics of the search. It's reverse engineering the 144 00:07:31,959 --> 00:07:34,839 drift path. It is if you know where the pieces 145 00:07:34,959 --> 00:07:39,399 ended up, and you have these advanced models of water movement, temperature, 146 00:07:39,439 --> 00:07:43,279 wind from that specific time period, you can greatly reduce 147 00:07:43,360 --> 00:07:46,240 the potential location of the original impact zone. 148 00:07:46,279 --> 00:07:49,360 Speaker 2: It reduces the variables in what is a truly complex. 149 00:07:48,920 --> 00:07:53,040 Speaker 1: Equation, and this leads directly to the most compelling quantitative factor. 150 00:07:53,160 --> 00:07:56,079 Sussi mentioned, the search zone. The dramatic reduction in the 151 00:07:56,079 --> 00:07:59,519 search zone. The original area of interest was described as 152 00:07:59,600 --> 00:08:00,720 about the size. 153 00:08:00,360 --> 00:08:02,040 Speaker 2: Of Florida, unbelievably huge. 154 00:08:02,079 --> 00:08:05,680 Speaker 1: And now thanks to this integrated application of AI analysis, 155 00:08:05,839 --> 00:08:10,319 swarm tech and this reanalyzed current modeling, the current specific 156 00:08:10,360 --> 00:08:13,319 area of high probability interest is only about the size 157 00:08:13,360 --> 00:08:14,000 of Connecticut. 158 00:08:14,040 --> 00:08:16,839 Speaker 2: I mean Florida to Connecticut. Debt reduction is not just massive, 159 00:08:16,920 --> 00:08:18,639 it's statistically game changing. 160 00:08:18,680 --> 00:08:19,240 Speaker 1: It has to be. 161 00:08:19,560 --> 00:08:23,680 Speaker 2: It means their confidence in the modeled location has increased. Geometrically, 162 00:08:24,199 --> 00:08:28,240 We're talking about replacing a broad low probability survey with 163 00:08:28,319 --> 00:08:31,439 an ultra focused high probability expedition. 164 00:08:31,680 --> 00:08:34,639 Speaker 1: Right, Susi said, This doesn't mean it's a less effective search. 165 00:08:34,679 --> 00:08:36,759 It means it's a more concise search. 166 00:08:36,679 --> 00:08:39,639 Speaker 2: And that concise targeting is everything when you're dealing with 167 00:08:39,679 --> 00:08:43,120 the expense and the time limitations of deep sea operations. 168 00:08:43,320 --> 00:08:47,080 Speaker 1: It really demonstrates that they believe the previous searches weren't 169 00:08:47,120 --> 00:08:50,960 necessarily failures of technology, but failures of data. They were 170 00:08:51,039 --> 00:08:52,679 just looking in the wrong place. 171 00:08:52,440 --> 00:08:55,120 Speaker 2: And by fixing the location modeling, they believe they are 172 00:08:55,159 --> 00:08:57,559 finally pointing the best technology at the right spot. 173 00:08:57,879 --> 00:09:01,399 Speaker 1: Okay, let's pivot entirely to the technical components now, because 174 00:09:01,840 --> 00:09:06,200 this is where Ocean Infinity's investment and Sussia's enthusiasm really originate. 175 00:09:06,399 --> 00:09:09,000 We need to dissect this swarm technology and the power 176 00:09:09,039 --> 00:09:09,519 it brings. 177 00:09:09,759 --> 00:09:13,039 Speaker 2: Absolutely, the difference between the old method and the new 178 00:09:13,440 --> 00:09:17,559 it really defines the efficiency gap. Historically, deep sea search 179 00:09:17,679 --> 00:09:21,080 relied on a large surface ship towing a device like 180 00:09:21,120 --> 00:09:24,360 a side scan sonar or a camera sled on a 181 00:09:24,480 --> 00:09:26,080 long miles long cable. 182 00:09:26,320 --> 00:09:29,960 Speaker 1: These were called towed underwater vehicles or tuvs. 183 00:09:30,360 --> 00:09:33,840 Speaker 2: That's right, and that system sounds inherently problematic when you're 184 00:09:33,879 --> 00:09:36,480 dealing with massive depth and unpredictable currents. 185 00:09:37,120 --> 00:09:41,480 Speaker 1: Incredibly slow and inefficient TV's covered a very narrow path, 186 00:09:41,879 --> 00:09:45,919 and the cable itself introduced latency and strain. If the 187 00:09:46,000 --> 00:09:49,120 surface ship moved just slightly off course because of weather 188 00:09:49,279 --> 00:09:52,759 or currents, the actual device three miles below could be 189 00:09:52,759 --> 00:09:57,120 a significantly displaced That compromises the integrity of the whole 190 00:09:57,159 --> 00:09:57,879 survey line. 191 00:09:57,919 --> 00:09:59,559 Speaker 2: So you're just constantly fighting the environment. 192 00:10:00,320 --> 00:10:03,120 Speaker 1: You are forced into these repetitive, slow, back and forth 193 00:10:03,200 --> 00:10:06,080 mowing the lawn patterns with a constant risk of the 194 00:10:06,120 --> 00:10:08,759 cable getting snagged or tangled on rugged terrain. 195 00:10:08,919 --> 00:10:11,960 Speaker 2: And now we have the swarm, which completely changes that 196 00:10:12,039 --> 00:10:17,519 operational geometry. We're talking about true autonomous underwater vehicles AUVs. 197 00:10:17,519 --> 00:10:21,799 Speaker 1: Correct. The new method uses multiple AUV units underwater simultaneously, 198 00:10:22,039 --> 00:10:24,440 and they are not tethered to the surface ship except 199 00:10:24,440 --> 00:10:26,159 maybe for data upload when they surface. 200 00:10:26,279 --> 00:10:29,679 Speaker 2: And crucially, they are tracked and controlled by single robotic boat, 201 00:10:30,120 --> 00:10:32,799 a vessel with not even anybody on board. 202 00:10:32,919 --> 00:10:33,639 Speaker 1: That's incredible. 203 00:10:33,720 --> 00:10:37,159 Speaker 2: This mothership acts as the acoustic command center, managing the 204 00:10:37,159 --> 00:10:40,639 telemetry and the missions for a whole fleet of underwater robots. 205 00:10:40,720 --> 00:10:43,679 Speaker 1: So that robotic mothership is coordinating the movement and the 206 00:10:43,759 --> 00:10:47,720 data collection from a dozen or more independent detectors. What's 207 00:10:47,759 --> 00:10:51,279 the fundamental technical leap that allows for this synchronization in 208 00:10:51,360 --> 00:10:54,879 deep water because you know, traditional GPS doesn't work down there. 209 00:10:54,919 --> 00:11:00,000 Speaker 2: That is the core technical marvel navigation synchronization. As you said, 210 00:11:00,000 --> 00:11:03,960 the GPS signals cannot penetrate water, especially at miles of depth. 211 00:11:04,559 --> 00:11:08,519 AUVs have to rely on acoustic telemetry for approximate positioning 212 00:11:08,600 --> 00:11:11,279 relative to the mothership and other AUVs. 213 00:11:10,919 --> 00:11:12,440 Speaker 1: And sound is slow in water. 214 00:11:12,759 --> 00:11:15,960 Speaker 2: Sound travels extremely slowly in water, about fifteen hundred meters 215 00:11:16,000 --> 00:11:19,480 per second. The latency and the multipath echoes make real 216 00:11:19,559 --> 00:11:22,799 time centimeter level control just impossible, so. 217 00:11:22,759 --> 00:11:25,080 Speaker 1: You can't steer them like remote control cars. They have 218 00:11:25,120 --> 00:11:28,200 to operate based on pre programmed routes exactly. 219 00:11:28,919 --> 00:11:32,559 Speaker 2: The real technical game changer is the internal inertial navigation 220 00:11:32,639 --> 00:11:36,879 systems i ins within each AUV. They're equipped with highly 221 00:11:36,919 --> 00:11:41,639 sensitive accelerometers and gyroscopes that track every single movement from 222 00:11:41,639 --> 00:11:45,080 a known starting point with exceptional precision. 223 00:11:44,919 --> 00:11:47,960 Speaker 1: AH, so they're self aware of their position relatively speaking. 224 00:11:48,120 --> 00:11:51,559 Speaker 2: Right, the mothership sends periodic acoustic updates to prevent long 225 00:11:51,639 --> 00:11:54,679 term drift, but for the most part, the AUVs are 226 00:11:54,720 --> 00:11:57,120 self navigating. For the duration of their mission and how 227 00:11:57,159 --> 00:12:00,159 long is a mission. It's typically constrained by battery life, 228 00:12:00,200 --> 00:12:02,759 so usually twenty four to forty eight hours before they 229 00:12:02,799 --> 00:12:06,080 have to return for charging and data download. This autonomy 230 00:12:06,120 --> 00:12:09,320 is what allows for the simultaneous searching in multiple areas 231 00:12:09,399 --> 00:12:10,080 all at one time. 232 00:12:10,159 --> 00:12:13,000 Speaker 1: This dramatically improves the cost to coverage ratio and the speed, 233 00:12:13,399 --> 00:12:16,759 but speed is useless without clear vision. Let's talk about 234 00:12:16,799 --> 00:12:19,720 the resolution power of these new devices. Susie said, the 235 00:12:19,759 --> 00:12:22,399 resolution is thousands of times better than it was before. 236 00:12:23,440 --> 00:12:26,480 What does that exponential lead mean in practical terms for 237 00:12:26,519 --> 00:12:28,200 the sonar operator, Well. 238 00:12:28,679 --> 00:12:32,600 Speaker 2: We're moving from older lower frequency analog side scan sonar 239 00:12:33,080 --> 00:12:38,840 to sophisticated high frequency digital sensors, specifically synthetic aperture sonar 240 00:12:39,159 --> 00:12:43,000 or sast okay. Older sonar provided large area coverage but 241 00:12:43,240 --> 00:12:47,039 very low detail. It essentially told you there is amount 242 00:12:47,080 --> 00:12:48,120 of something here. 243 00:12:47,879 --> 00:12:51,200 Speaker 1: And that low detail created that ambiguity problem you mentioned earlier. 244 00:12:51,240 --> 00:12:55,879 Speaker 2: Precisely, the older systems forced operators to constantly ask is 245 00:12:55,919 --> 00:12:58,000 it a rock? Is it a boat? Is it a 246 00:12:58,000 --> 00:13:01,120 crashed airplane? Right? And if the target was ambiguous, the 247 00:13:01,279 --> 00:13:04,840 entire expensive operation had to pause. Haul the AUV backup, 248 00:13:05,080 --> 00:13:07,679 maybe switch out the sensors and then perform a slow 249 00:13:07,799 --> 00:13:11,440 close range investigation. This just introduced massive downtime and. 250 00:13:11,440 --> 00:13:14,240 Speaker 1: Cost, so the high resolution must fundamentally reduce the number of. 251 00:13:14,159 --> 00:13:18,360 Speaker 2: False positives absolutely. SAS uses complex signal processing to generate 252 00:13:18,399 --> 00:13:22,360 acoustic imagery that approaches photolike clarity. It can resolve details 253 00:13:22,360 --> 00:13:24,919 out of centimeters, so operators can now look at the 254 00:13:24,919 --> 00:13:29,440 sonar return and often immediately distinguish a naturally eroded rock 255 00:13:29,519 --> 00:13:33,320 formation from a geometrically structured object like a piece of 256 00:13:33,360 --> 00:13:38,399 fuselage or an engine housing. This high res capability eliminates 257 00:13:38,440 --> 00:13:41,279 the need for those time consuming investigative passes over every 258 00:13:41,279 --> 00:13:42,080 little anomaly. 259 00:13:42,320 --> 00:13:44,879 Speaker 1: So let's connect this vision to the brain the AI 260 00:13:45,000 --> 00:13:48,440 system analyzing the data. The data comes in as sonar, 261 00:13:48,679 --> 00:13:53,039 essentially pixels of different colors representing different densities and acoustic shadows. 262 00:13:53,759 --> 00:13:56,799 How does AI leverage this ultra high resolution stream. 263 00:13:57,279 --> 00:14:00,399 Speaker 2: The AI is applied to analyze this vast stream of data, 264 00:14:00,679 --> 00:14:02,960 and its power lies in its ability to look at 265 00:14:03,000 --> 00:14:06,279 every single data point and flag potential targets that might 266 00:14:06,279 --> 00:14:08,600 have been missed due to the limits of human perception. 267 00:14:08,799 --> 00:14:10,879 Speaker 1: The stale is just too big for humans alone. 268 00:14:11,000 --> 00:14:14,240 Speaker 2: Think of it. The AUV swarm is collecting terabytes of 269 00:14:14,279 --> 00:14:17,480 SaaS data across an area the size of Connecticut. No 270 00:14:17,679 --> 00:14:20,960 human team could review that comprehensively in the operational timeframe. 271 00:14:21,159 --> 00:14:23,960 Speaker 1: We talked about AI filtering noise, but let's get specific. 272 00:14:24,399 --> 00:14:27,919 What signatures is the AI trained to look for? Is 273 00:14:27,960 --> 00:14:30,679 it trained on wreckage from other plane crashes or does 274 00:14:30,720 --> 00:14:33,480 it look for generic man made materials both? 275 00:14:33,679 --> 00:14:37,360 Speaker 2: Really, it is trained on known debris imagery, but more importantly, 276 00:14:37,399 --> 00:14:41,399 it looks for angularity and material contrast. Natural formations tend 277 00:14:41,480 --> 00:14:44,039 to be smooth and curved by erosion over time. 278 00:14:44,159 --> 00:14:44,440 Speaker 1: Okay. 279 00:14:44,919 --> 00:14:49,080 Speaker 2: Aviation wreckage, however, presents hard edges, repeating geometric patterns like 280 00:14:49,120 --> 00:14:53,480 window frames or wing ribs, and density contrasts specific to titanium, 281 00:14:53,639 --> 00:14:57,639 lunum alloys and composite materials. The AI can process those 282 00:14:57,639 --> 00:15:01,080 specific signatures far faster and more reliable than the human 283 00:15:01,120 --> 00:15:03,480 eye scanning a sonar scroll for hours on end. 284 00:15:03,840 --> 00:15:07,559 Speaker 1: This is a powerful synthesis. Then the swarm provides unprecedented 285 00:15:07,679 --> 00:15:10,960 coverage and resolution, and the AI provides unprecedented analysis and 286 00:15:11,039 --> 00:15:14,600 quality control. David Gallow's assessment that ocean infinity came out 287 00:15:14,639 --> 00:15:16,960 of nowhere and gangbusters and are now leading the industry 288 00:15:16,960 --> 00:15:20,279 for deep sea surveys. That just underscores the credibility of 289 00:15:20,320 --> 00:15:24,720 this technological stack. They have clearly solved several major operational bottlenecks. 290 00:15:24,960 --> 00:15:28,039 Speaker 2: They have they have minimized the risk that they either 291 00:15:28,080 --> 00:15:30,279 missed the wreck in the past or will miss it 292 00:15:30,360 --> 00:15:34,600 in the present due to technological ambiguity or human error. Now, 293 00:15:34,720 --> 00:15:38,559 the entire mission hinges on the accuracy of that initial 294 00:15:38,639 --> 00:15:41,159 data modeling the location of the haystack itself. 295 00:15:41,360 --> 00:15:43,679 Speaker 1: Moving on, let's shift from the tools of the search 296 00:15:44,000 --> 00:15:47,639 to the physical clues. They've already gathered the small wreckage 297 00:15:47,639 --> 00:15:51,159 pieces that washed ashore thousands of miles away. They provided 298 00:15:51,240 --> 00:15:54,919 critical forensic insight into the final moments of that flight. 299 00:15:55,159 --> 00:15:58,080 Speaker 2: Yeah, this is where the story gets really granular. We're 300 00:15:58,080 --> 00:16:02,720 talking about two primary catagories of located wreckage. The flaperone, 301 00:16:02,720 --> 00:16:06,240 which is a crucial movable control surface on the trailing 302 00:16:06,320 --> 00:16:08,639 edge of the wing, and the leading edges some flight 303 00:16:08,679 --> 00:16:11,240 control areas. All of these were confirmed to be from 304 00:16:11,320 --> 00:16:14,840 MH three seventy. These pieces traveled for thousands of miles, 305 00:16:15,200 --> 00:16:17,440 but their physical condition is like an open book on 306 00:16:17,480 --> 00:16:19,080 the plane's final dynamic. 307 00:16:18,720 --> 00:16:21,159 Speaker 1: State, and the analysis of this debris led to a 308 00:16:21,159 --> 00:16:25,320 conclusion that completely reframed the narrative of the plane's final descent. 309 00:16:26,080 --> 00:16:28,759 What did the debris tell the forensic investigators? 310 00:16:29,039 --> 00:16:32,879 Speaker 2: The critical finding from the debris analysis, particularly from the flaperone, 311 00:16:33,159 --> 00:16:35,919 was that it came off in flight, okay, And this 312 00:16:35,960 --> 00:16:40,240 is a crucial distinction that separates, say, a controlled water landing, 313 00:16:40,480 --> 00:16:41,759 from a high energy crash. 314 00:16:42,000 --> 00:16:45,440 Speaker 1: Help us understand the forensic difference between separation in flight 315 00:16:45,639 --> 00:16:49,000 and separation on impact. For the average person, I think 316 00:16:49,039 --> 00:16:52,000 a piece breaking off is just a piece breaking off, right. 317 00:16:52,840 --> 00:16:55,480 Speaker 2: The difference lies in the nature of the fracture and 318 00:16:55,559 --> 00:16:59,120 the wear marks. If an aircraft descends slowly and impacts 319 00:16:59,120 --> 00:17:02,759 the water relatively le gently a controlled ditching, which would 320 00:17:02,799 --> 00:17:03,879 happen near stall. 321 00:17:03,679 --> 00:17:05,839 Speaker 1: Speed like the miracle on the Hudson. 322 00:17:05,559 --> 00:17:08,519 Speaker 2: Exactly like that, the components would break off due to 323 00:17:08,559 --> 00:17:12,039 the sudden hydraulic pressure of hitting the water. This results 324 00:17:12,039 --> 00:17:16,440 in characteristic shear marks, buckling, and fragmentation consistent with a 325 00:17:16,480 --> 00:17:18,000 forward moving water impact. 326 00:17:18,160 --> 00:17:20,200 Speaker 1: And what did they actually find on the MH three 327 00:17:20,240 --> 00:17:21,160 seventy flapperone. 328 00:17:21,640 --> 00:17:25,400 Speaker 2: They found evidence that suggests aerodynamic overload for a component 329 00:17:25,480 --> 00:17:28,240 like a flap erone separation in flight means the force 330 00:17:28,400 --> 00:17:31,480 the dynamic pressure acting on the wing structure exceeded the 331 00:17:31,559 --> 00:17:33,440 ultimate structural limits before it even. 332 00:17:33,319 --> 00:17:35,440 Speaker 1: Touched the water, so it was torn off by the 333 00:17:35,480 --> 00:17:36,200 air itself. 334 00:17:36,240 --> 00:17:39,279 Speaker 2: The component was likely torn cleanly from its mounting points 335 00:17:39,599 --> 00:17:43,200 due to massive forces, perhaps exceeding the design limit of 336 00:17:43,240 --> 00:17:47,680 the plane. This implies the aircraft was moving violently well 337 00:17:47,720 --> 00:17:50,440 above safe operational speed during its final moments. 338 00:17:50,640 --> 00:17:54,519 Speaker 1: That single piece of analysis proving an aerodynamic overload separation 339 00:17:54,960 --> 00:17:58,960 changes everything about the final trajectory. It suggests the plane 340 00:17:59,119 --> 00:18:03,720 entered an uncontrol rolled high speed dive or spiral. It 341 00:18:03,759 --> 00:18:06,279 was a high energy event in the air, not a 342 00:18:06,359 --> 00:18:09,160 gentle end to a controlled glide exactly. 343 00:18:09,640 --> 00:18:12,480 Speaker 2: This evidence strongly suggests the aircraft made a rapid descent 344 00:18:12,720 --> 00:18:16,240 and it came apart somewhat in flight, specifically, at least 345 00:18:16,279 --> 00:18:18,960 the flight controls did. This implies the plane was not 346 00:18:19,000 --> 00:18:22,319 being controlled effectively or at all in its final moments, 347 00:18:22,319 --> 00:18:24,559 and it achieved a terminal velocity or an angle of 348 00:18:24,599 --> 00:18:27,039 attack that simply ripped the surface right off. 349 00:18:27,119 --> 00:18:29,759 Speaker 1: So let's connect these forensic dots to the search area. 350 00:18:30,000 --> 00:18:32,839 If the controls detached in flight during a rapid descent. 351 00:18:33,240 --> 00:18:36,039 What does that suggest about the plane's final trajectory and 352 00:18:36,119 --> 00:18:39,480 speed after its last known handshake with the satellite. 353 00:18:39,759 --> 00:18:43,160 Speaker 2: Well, it solidifies the theory that the plane entered a steep, 354 00:18:43,240 --> 00:18:46,440 high speed dive. The impact site will not be a 355 00:18:46,480 --> 00:18:50,319 relatively intact fuselage resting gently on the seabed. No, they 356 00:18:50,319 --> 00:18:53,759 are looking for a highly scattered wreckage field. Though the 357 00:18:53,799 --> 00:18:56,599 heaviest pieces the engines, the landing gear, the main wing 358 00:18:56,640 --> 00:18:59,880 spars in the fuselage center section, they would fall most 359 00:19:00,079 --> 00:19:03,519 vertically and concentrate at the primary impact point. 360 00:19:03,599 --> 00:19:06,720 Speaker 1: This forensic detail is vital because it informs the modeling 361 00:19:06,839 --> 00:19:10,279 used by the AI. Knowing a plane likely impacted at 362 00:19:10,319 --> 00:19:13,960 high velocity allows them to model the likely dispersion radius 363 00:19:14,000 --> 00:19:17,119 of the wreckage, which helps them define that Connecticut sized 364 00:19:17,160 --> 00:19:20,319 box more accurately. It confirms they are looking for a 365 00:19:20,400 --> 00:19:21,279 violent crash site. 366 00:19:21,319 --> 00:19:24,000 Speaker 2: The debris analysis helps narrow the line of the final 367 00:19:24,039 --> 00:19:27,720 trajectory through the current modeling, and the crash dynamics inform 368 00:19:27,759 --> 00:19:30,359 the necessary size of the immediate target zone on the 369 00:19:30,400 --> 00:19:33,640 ocean floor. The fact that the search area has shrunk 370 00:19:34,000 --> 00:19:37,000 suggests they have successfully modeled the last known movements and 371 00:19:37,039 --> 00:19:40,519 the crash site with high precision, even given the violence 372 00:19:40,559 --> 00:19:41,480 of the likely impact. 373 00:19:41,839 --> 00:19:45,200 Speaker 1: Now let's inject the necessary caution and reality check into 374 00:19:45,240 --> 00:19:49,480 this highly technological discussion. We move to David Gallo, the 375 00:19:49,480 --> 00:19:54,000 oceanographer who has personally participated in these landmark deep sea expeditions, 376 00:19:54,519 --> 00:19:58,119 and he provides the counter perspective the optimism of Susi 377 00:19:58,279 --> 00:20:02,359 meets the sheer, unyielded difficulty of the deep ocean floor. 378 00:20:03,039 --> 00:20:06,359 Speaker 2: Gallo's perspective is absolutely vital because he reminds us that 379 00:20:06,759 --> 00:20:09,319 no matter how advanced your instruments are, you still have 380 00:20:09,359 --> 00:20:11,759 to deal with the most challenging and least mapped environment 381 00:20:11,839 --> 00:20:12,200 on Earth. 382 00:20:12,240 --> 00:20:14,079 Speaker 1: He had a great analogy for it, he did. 383 00:20:14,279 --> 00:20:16,960 Speaker 2: He framed the search effort beautifully, saying, it's like an 384 00:20:17,039 --> 00:20:20,039 orchestra that requires the right team, the right instruments, and 385 00:20:20,079 --> 00:20:21,839 the right technique the music. 386 00:20:21,960 --> 00:20:26,720 Speaker 1: But the true existential challenge remains external to the technology itself. 387 00:20:26,799 --> 00:20:29,680 Speaker 2: Precisely determining if they are looking in the right haystack. 388 00:20:30,079 --> 00:20:33,240 Gallo understands the difference between having the best tools and 389 00:20:33,319 --> 00:20:36,119 knowing precisely where to aim them. He knows the ocean 390 00:20:36,160 --> 00:20:39,279 is vast enough to hide something permanently, even a Boeing 391 00:20:39,319 --> 00:20:40,200 seven seven seven. 392 00:20:40,799 --> 00:20:44,960 Speaker 1: Let's detail the immense difficulties of deep underwater searching. Because 393 00:20:45,000 --> 00:20:48,440 our listeners might hear about robotic swarms and high resolutions 394 00:20:48,440 --> 00:20:51,680 sonar and just assume the challenge has been eliminated. The 395 00:20:51,759 --> 00:20:53,200 reality is far harsher. 396 00:20:53,640 --> 00:20:57,319 Speaker 2: It is. The difficulties start with depth. The search area 397 00:20:57,359 --> 00:21:01,240 located in the remote southern Indian Ocean, the Abyssle plane, 398 00:21:01,440 --> 00:21:03,440 where the floor can range from an average of two 399 00:21:03,480 --> 00:21:05,880 and a half miles deep up to four or five 400 00:21:05,960 --> 00:21:10,000 miles stack. Operating robotic equipment under the pressure of several 401 00:21:10,039 --> 00:21:14,640 thousand pounds per square inch requires incredibly robust engineering, not 402 00:21:14,759 --> 00:21:18,240 to mention the challenge of navigation accuracy at those distances. 403 00:21:18,720 --> 00:21:21,440 Speaker 1: And then there is the topography, which Gallo said makes 404 00:21:21,440 --> 00:21:23,079 even the rocky mountains look mild. 405 00:21:23,319 --> 00:21:25,960 Speaker 2: That's right. He stated that roughly one third of the 406 00:21:26,000 --> 00:21:30,039 ocean floor is incredibly mountainous, potentially more steep and rugged 407 00:21:30,079 --> 00:21:32,960 than the Rockies or the Himalayas. Wow, we are talking 408 00:21:33,000 --> 00:21:38,359 about abyssle hills, fracture zones, sheer cliffs, all in perpetual darkness. 409 00:21:38,680 --> 00:21:42,480 Speaker 1: This complexity must pose specific problems for AUVs, doesn't it. 410 00:21:42,960 --> 00:21:45,880 If the AUV is programmed to follow a set altitude 411 00:21:45,920 --> 00:21:50,799 above the seabed, volatile topography must create navigational and data challenges. 412 00:21:50,920 --> 00:21:54,400 Speaker 2: It creates huge risks. The AAV has to constantly use 413 00:21:54,400 --> 00:21:57,680 its forward looking sonar and its ions to adjust its 414 00:21:57,720 --> 00:22:02,440 altitude to avoid a collision, ensumes precious battery life. If 415 00:22:02,440 --> 00:22:05,640 the terrain changes too rapidly a sudden escarpment or a 416 00:22:05,640 --> 00:22:09,279 deep canyon, the AUV might either miss the wall completely, 417 00:22:09,519 --> 00:22:12,079 which is a crash risk, or it might fly too high, 418 00:22:12,400 --> 00:22:15,079 creating a shadow zone on the slope where wreckage could 419 00:22:15,119 --> 00:22:16,119 hide perfectly. 420 00:22:15,759 --> 00:22:18,319 Speaker 1: From the sonar ah an acoustic shadow. 421 00:22:18,480 --> 00:22:20,359 Speaker 2: You can't just fly a straight line over a region 422 00:22:20,359 --> 00:22:21,680 that is steeper than the MLAIS. 423 00:22:21,920 --> 00:22:25,319 Speaker 1: That problem of acoustic shadows is significant If the debris 424 00:22:25,400 --> 00:22:28,920 is large, but it's resting on a steep, acoustically shielded slope, 425 00:22:29,200 --> 00:22:31,480 even the best sonar might miss it. If the path 426 00:22:31,519 --> 00:22:34,160 doesn't provide the perfect angle of incidents. 427 00:22:33,680 --> 00:22:37,000 Speaker 2: Which means you need overlapping coverage and multiple passes, which 428 00:22:37,039 --> 00:22:39,160 burns time and operational resources. 429 00:22:39,440 --> 00:22:44,000 Speaker 1: And that difficulty is compounded by surface factors. Gallo reminds 430 00:22:44,079 --> 00:22:47,240 us that even with robotic control boats, the mothership still 431 00:22:47,279 --> 00:22:50,599 has to deal with weather and strong ocean currents. While 432 00:22:50,640 --> 00:22:54,960 the AUVs are autonomous, their navigation is periodically corrected by 433 00:22:55,000 --> 00:22:58,079 acoustic signals from the surface. So if the surface vessel 434 00:22:58,119 --> 00:23:01,440 is being buffeted by a major storm, the navigational precision 435 00:23:01,480 --> 00:23:04,519 of the entire swarm below, which is already challenging at 436 00:23:04,559 --> 00:23:07,480 three or four miles deep, it suffers. 437 00:23:07,160 --> 00:23:10,680 Speaker 2: You risk less accurate survey lines. And then there's this 438 00:23:10,759 --> 00:23:14,960 year operational difficulty of the entire process, which is just humbling. 439 00:23:15,480 --> 00:23:19,640 Gallo summarized it as the logistical nightmare of lowering an 440 00:23:19,680 --> 00:23:21,960 instrument or instruments to the bottom of the ocean and 441 00:23:22,000 --> 00:23:24,839 putting them to work. Yeah, keeping a fleet of complex 442 00:23:24,960 --> 00:23:28,440 robots running efficiently, safely, and continuously for weeks on end 443 00:23:28,480 --> 00:23:32,599 in salt water under crushing pressure. It's a testament to 444 00:23:32,599 --> 00:23:35,799 the engineers. But it's also a persistent source of potential failure. 445 00:23:35,920 --> 00:23:38,680 Speaker 1: This is why dallas skepticism is so important. He is 446 00:23:38,720 --> 00:23:42,440 fundamentally challenging the input data while he hopes and prays 447 00:23:42,559 --> 00:23:45,480 for a find. He noted that he does not personally 448 00:23:45,519 --> 00:23:48,359 see where the new information has come from. That allowed 449 00:23:48,359 --> 00:23:51,920 the search area to shrink so dramatically. The underlying satellite 450 00:23:52,000 --> 00:23:56,680 data the in Marsat pings, that hasn't changed since twenty fourteen. 451 00:23:56,599 --> 00:24:00,039 Speaker 2: And that's a powerful challenge. Susie is confident in the 452 00:24:00,079 --> 00:24:03,960 AI and the revised current modeling. Gallo is asking is 453 00:24:04,000 --> 00:24:07,640 the model actually correct or is the confidence misplaced? 454 00:24:08,000 --> 00:24:10,319 Speaker 1: The risk is that Ocean Infinity is applying the most 455 00:24:10,359 --> 00:24:13,440 advanced tools in the world to a precisely wrong location. 456 00:24:13,920 --> 00:24:17,039 Speaker 2: We have to respect both positions. Susi is driven by 457 00:24:17,039 --> 00:24:20,440 the potential of innovation and the power of computational analysis. 458 00:24:21,079 --> 00:24:23,640 Gallo is driven by the hard won experience of the 459 00:24:23,640 --> 00:24:27,759 Ocean's difficulty and the inherent uncertainties of long distance acoustic 460 00:24:27,799 --> 00:24:30,039 data model. The tension is real, it is, and it 461 00:24:30,039 --> 00:24:33,319 won't be resolved until the AUVs send back definitive imagery. 462 00:24:33,799 --> 00:24:36,559 Speaker 1: Let's turn our final analytical focus to the prize and 463 00:24:36,599 --> 00:24:39,599 the payoff. Assuming Ocean Infinity is in the right haystack 464 00:24:39,720 --> 00:24:43,279 and their swarm technology successfully locates the main wreckage, we 465 00:24:43,319 --> 00:24:46,400 are talking about securing the critical data necessary to solve 466 00:24:46,400 --> 00:24:47,680 the why of this mystery. 467 00:24:47,880 --> 00:24:50,960 Speaker 2: The prize, undoubtedly is the recovery of the flight data 468 00:24:51,000 --> 00:24:54,599 recorder the FDR, and the cockpit voice recorder the CVR, 469 00:24:54,680 --> 00:24:57,759 the so called black boxes. Right finding the main body 470 00:24:57,759 --> 00:25:01,240 of the aircraft is essential because those devices they're designed 471 00:25:01,240 --> 00:25:05,119 to survive extreme impact, and they should still be intact 472 00:25:05,160 --> 00:25:08,400 and likely attached deep within the main airframe structure. 473 00:25:08,480 --> 00:25:10,960 Speaker 1: But after a decade of searching, the years of waiting, 474 00:25:11,119 --> 00:25:14,400 does that time span compromise the data? Are those black 475 00:25:14,440 --> 00:25:18,599 boxes still functional or more accurately, are they still holding 476 00:25:18,640 --> 00:25:19,160 their data? 477 00:25:19,640 --> 00:25:23,119 Speaker 2: According to David Suzi, the longevity of these recorders is robust. 478 00:25:23,799 --> 00:25:25,759 He confirmed that they are designed to last ten to 479 00:25:25,799 --> 00:25:29,440 fifteen years and potentially much longer in that stable, cold, 480 00:25:29,680 --> 00:25:32,920 deep ocean environment. Okay, Crucially, the data is stored on 481 00:25:33,000 --> 00:25:36,440 EPROM chips, which are engineered to hold data securely for 482 00:25:36,480 --> 00:25:40,359 that duration, even when submerged under immense pressure and exposed 483 00:25:40,359 --> 00:25:41,440 to corrosive saltwater. 484 00:25:41,640 --> 00:25:45,359 Speaker 1: EPROM irascible programmable read only memory. This is key because 485 00:25:45,359 --> 00:25:48,640 it's a non volatile form of storage. Why does nonvolatility 486 00:25:48,680 --> 00:25:50,880 matter so much for deep sea data retrieval? 487 00:25:51,200 --> 00:25:54,400 Speaker 2: It means the chips do not need continuous power to 488 00:25:54,400 --> 00:25:57,759 retain the information. The data is physically stored in the 489 00:25:57,799 --> 00:25:58,920 circuit structure itself. 490 00:25:59,039 --> 00:26:00,079 Speaker 1: Okay, so it's bake. 491 00:26:00,519 --> 00:26:03,160 Speaker 2: It's baked in. The immense pressure and the lack of 492 00:26:03,200 --> 00:26:06,880 oxygen in the deep water actually aid preservation by slowing 493 00:26:06,920 --> 00:26:12,359 down typical degradation processes like microbial or chemical corrosion that 494 00:26:12,440 --> 00:26:16,240 might affect other components. If the armoured casing remains intact, 495 00:26:16,759 --> 00:26:19,599 the e Prompt chips should survive for decades. 496 00:26:19,920 --> 00:26:22,680 Speaker 1: So the challenge is not data integrity, but the logistics 497 00:26:22,680 --> 00:26:25,880 of physically retrieving the boxes and successfully interfacing with the 498 00:26:26,000 --> 00:26:28,880 chips after all this time exactly. And the black boxes 499 00:26:28,880 --> 00:26:31,079 don't just tell us that the plane crashed. They contain 500 00:26:31,279 --> 00:26:35,400 two fundamentally different categories of information that together paint the 501 00:26:35,440 --> 00:26:39,960 full picture of the catastrophe. What distinguishes the CVR from 502 00:26:39,960 --> 00:26:42,720 the FDR in terms of the insight they give well. 503 00:26:42,799 --> 00:26:45,720 Speaker 2: The FDR the flight data recorder. It records thousands of 504 00:26:45,720 --> 00:26:51,359 specific operational parameters airspeed, altitude, heading, engine performance, the positions 505 00:26:51,359 --> 00:26:54,519 of the control surfaces, autopilot commands, the nuts and bolts, 506 00:26:54,720 --> 00:26:59,000 all the nuts and bolts. This data, logged second by second, 507 00:26:59,279 --> 00:27:03,319 tells us the sise, physical trajectory and performance status of 508 00:27:03,359 --> 00:27:06,119 the seven seven seven leading up to impact. It's the 509 00:27:06,160 --> 00:27:08,519 definitive engineering report of what the plane was. 510 00:27:08,480 --> 00:27:11,200 Speaker 1: Doing, and the CVR, the cockpit voice. 511 00:27:10,960 --> 00:27:13,279 Speaker 2: For CVR is the human element. It records the final 512 00:27:13,319 --> 00:27:17,119 two hours of audio pilot voices, air traffic control communication, 513 00:27:17,640 --> 00:27:21,960 ambient sounds in the cockpit, and crucial operational alarms or warnings. 514 00:27:22,599 --> 00:27:25,319 If there was a struggle for control, a fire, a 515 00:27:25,359 --> 00:27:28,440 sudden decompression, or a deliberate action by the pilot, the 516 00:27:28,440 --> 00:27:31,839 CDR captures the human reaction and the timeline of those events. 517 00:27:31,960 --> 00:27:33,880 Speaker 1: So this is the difference between knowing where it is 518 00:27:33,960 --> 00:27:35,039 and knowing what happened. 519 00:27:35,119 --> 00:27:35,519 Speaker 2: That's it. 520 00:27:35,720 --> 00:27:38,960 Speaker 1: Did the pilot intentionally crash the plane as some theories suggest. 521 00:27:39,240 --> 00:27:41,799 Was there an electrical fire that incapacitated the crew and 522 00:27:41,880 --> 00:27:45,359 led to this chaotic, uncontrolled flight. Was it a mechanical 523 00:27:45,359 --> 00:27:48,920 failure that just spiraled catastrophically out of control. The data 524 00:27:49,079 --> 00:27:52,160 locked within those recorders holds the definitive answers. 525 00:27:51,880 --> 00:27:54,880 Speaker 2: And for the families, this means moving beyond a decade 526 00:27:54,880 --> 00:28:00,119 of painful speculation. For global aviation safety, that data is priceless. 527 00:28:00,640 --> 00:28:03,359 It prevents this event from being categorized merely as an 528 00:28:03,440 --> 00:28:08,440 unsolvable anomaly and transforms it into a solvable, definitive safety 529 00:28:08,480 --> 00:28:12,640 case that can inform future aircraft design and training protocols. 530 00:28:12,359 --> 00:28:16,359 Speaker 1: And that ultimately is the true payoff of Ocean Infinity's 531 00:28:16,519 --> 00:28:21,200 incredibly high tech, high stakes endeavor. They aren't just searching 532 00:28:21,240 --> 00:28:24,160 for a ghost plane. They're searching for the definitive cause 533 00:28:24,200 --> 00:28:27,039 of a catastrophic event, one that fundamentally changed how we 534 00:28:27,200 --> 00:28:31,440 view everything from satellite tracking to air travel safety procedures. 535 00:28:31,839 --> 00:28:34,519 So let's synthesize the two viewpoints we've explored in this 536 00:28:34,680 --> 00:28:37,160 edition of Thrilling Threads. On one side, we have the 537 00:28:37,240 --> 00:28:41,720 highly technological optimism championed by David Sussi. You've got AI 538 00:28:41,920 --> 00:28:46,160 filtering old data and identifying subtle patterns, swarm robotics increasing 539 00:28:46,160 --> 00:28:48,960 search efficiency by factors of thousands, and a search area 540 00:28:49,000 --> 00:28:50,839 reduced from the size of Florida to the size of 541 00:28:50,839 --> 00:28:54,920 Connecticut based on sophisticated current modeling and forensic evidence. The 542 00:28:54,960 --> 00:28:57,680 technology has never been better positioned to succeed. 543 00:28:57,920 --> 00:29:00,400 Speaker 2: And on the other side, we have the experience against 544 00:29:00,480 --> 00:29:03,880 scientific caution of David Gallo. The ocean is still three 545 00:29:03,880 --> 00:29:07,359 to five miles deep, the bottom is rugged, challenging the 546 00:29:07,359 --> 00:29:08,880 AUVs at every turn. 547 00:29:08,960 --> 00:29:11,359 Speaker 1: And the fundamental question remains, is. 548 00:29:11,279 --> 00:29:14,160 Speaker 2: The initial data model, regardless of how refined it is, 549 00:29:14,440 --> 00:29:17,640 pointing to the right haystack. The optimism is warranted by 550 00:29:17,680 --> 00:29:20,960 the tools, but the pessimism is warranted by the environment 551 00:29:21,000 --> 00:29:23,000 and the history of this specific failure. 552 00:29:23,079 --> 00:29:25,720 Speaker 1: And here's where it gets really interesting for you, the listener. 553 00:29:26,200 --> 00:29:29,440 The search area is reduced, the technology is exponentially better, 554 00:29:29,720 --> 00:29:31,960 and the hope is high that the recovery of the 555 00:29:32,000 --> 00:29:33,440 black boxes is imminent. 556 00:29:33,519 --> 00:29:34,920 Speaker 2: It feels closer than ever. 557 00:29:35,079 --> 00:29:38,160 Speaker 1: If they succeed, we will finally have the definitive story 558 00:29:38,200 --> 00:29:41,599 of MH three seventy. This new knowledge, however, might come 559 00:29:41,640 --> 00:29:45,240 with painful, excruciating clarity for the families about the exact, 560 00:29:45,440 --> 00:29:49,799 potentially horrific circumstances of the final moments recorded on the CVR. 561 00:29:50,079 --> 00:29:53,079 Speaker 2: The painful uncertainty is lasted a decade, but it allowed 562 00:29:53,119 --> 00:29:56,359 for speculation and perhaps for maintaining a sliver of hope 563 00:29:56,400 --> 00:29:59,559 that the circumstances were less terrifying than a high speed 564 00:29:59,640 --> 00:30:00,720 aired n breakup. 565 00:30:00,839 --> 00:30:02,599 Speaker 1: So here's the question we want to leave you with. 566 00:30:03,279 --> 00:30:06,480 Given the choice between the current decade of uncertainty and 567 00:30:06,519 --> 00:30:11,000 a potentially horrific but definitive answer, locked inside those resilient 568 00:30:11,039 --> 00:30:14,359 e prom chips. What do you think the global community 569 00:30:14,400 --> 00:30:17,880 needs most closure, no matter how painful or the continuation 570 00:30:18,000 --> 00:30:20,519 of the painful silence that has persisted for ten. 571 00:30:20,440 --> 00:30:23,319 Speaker 2: Years, the knowledge is waiting at the bottom of the sea. 572 00:30:23,440 --> 00:30:25,960 Speaker 1: Share your stand on the implications of this final search 573 00:30:26,000 --> 00:30:28,240 and the comments. Thank you for diving deep with us 574 00:30:28,240 --> 00:30:29,200 on Thrilling Threads.