I Built a God’s Eye View of Time. The Internet Went Wild. — Transcript
Full transcript
- 0:00I turned this mama deer and her babies
- 0:02walk into a 4D reconstruction and I
- 0:04started wondering, is this how higher
- 0:06dimensional beings see us? Every moment
- 0:09all at once. I put it online and holy
- 0:12crap, it went viral on every platform
- 0:15I've got. Some people saw computer
- 0:17vision, 4D reconstruction from 2D
- 0:19videos. Others went full philosophical
- 0:22and honestly, I get both reactions. But
- 0:24the deer, my friends, is just the
- 0:26beginning. What if you could put all
- 0:28your photos and videos back when and
- 0:31where they happened and explore those
- 0:33memories from a god's eye view. And what
- 0:36if you could build this time machine not
- 0:37only for your camera roll, but the whole
- 0:39damn planet?
- 0:41Welcome to the spatial memory palace.
- 0:44Let's get into it.
- 0:45So this is a clip I recorded of the mama
- 0:47deer and her baby, right? Typically, we
- 0:48experience life sequentially. It's like
- 0:50flipping through a book, one frame after
- 0:52another, but the previous one
- 0:53disappears, but you see motion. In this
- 0:56case, I'm leaving every step behind in
- 0:583D. Each of these poses is coming from a
- 1:01different moment in the video. So when
- 1:02you start putting it together, you start
- 1:05getting this dimensional shape that you
- 1:06can look around. Now by 4D, I mean the
- 1:09three dimensions of space, X, Y, and Z,
- 1:12and then time as the fourth dimension.
- 1:14So I can take these video frames and use
- 1:16the latest state-of-the-art models to
- 1:18reconstruct their structure in 3D. And
- 1:21instead of playing back those moments
- 1:22one after another, I can start leaving
- 1:24them in a scene and move around them.
- 1:26It's kind of like onion skinning reality
- 1:28if you're familiar with the animation
- 1:30analogy. Now people have been stacking
- 1:32video frames to visualize time for many
- 1:34years. So when I saw this cube clip
- 1:36going around, it made me want to try it
- 1:38with real 3D moments, not just 2D
- 1:41frames. So of course, I had to try it
- 1:43with myself and this is where Kurt
- 1:44Vonnegut comes in. Here you can see me
- 1:46moving through the world, right? Like
- 1:47leaving a 3D snapshot of every single
- 1:50moment exactly where it happened. Now
- 1:52with the deer, it feels like a cool
- 1:53visual experiment, but when you see
- 1:55yourself, there's something a little bit
- 1:57more existential that comes to mind. And
- 1:59Kurt Vonnegut basically nails it. So, in
- 2:02Kurt Vonnegut's story, there's this
- 2:03alien race that don't see humans as
- 2:06two-legged creatures. They see them as a
- 2:08great millipede. The baby legs on one
- 2:10end and old people legs on the other.
- 2:12And my god, the reactions are super
- 2:14profound. Gabriella brings up this point
- 2:16that we're sort of like caterpillars
- 2:17burrowing through space, right? That's
- 2:19kind of what it feels like when you look
- 2:21at this deer kind of tunneling through
- 2:23space. Dr. Kai Bulli Neale of it, too. I
- 2:25mean, just take the example of you from
- 2:27morning till the evening. We view these
- 2:29as sequential moments. What if you could
- 2:31see it all at once? Now, of course, if
- 2:33you step outside of time and you could
- 2:35see it all at once, it begs the
- 2:36question, like, do we have free will or
- 2:38is everything preordained and
- 2:40predestined? Like Lester King brings up
- 2:42the analogy, like higher dimensions
- 2:43exist like a book. We experience one
- 2:46page at a time, but the book already
- 2:48exists in its entirety. So, for a
- 2:50theoretical higher-dimensional being
- 2:52looking at us, basically, they've got
- 2:55all the plot twists. They know the
- 2:56ending. Now, Areef brings up this idea
- 2:58of angels who record like cosmic scribes
- 3:01that are building and maintaining this
- 3:03audit trail of everything you've done
- 3:05and potentially even everything you
- 3:06thought. Like Vedic traditions and Sikh
- 3:08traditions, for example, have this idea
- 3:10of Chitragupta, literally cosmic scribes
- 3:13that record everything that you do in
- 3:15the physical world as well as things
- 3:16that you think. But I mean, look, I
- 3:17don't care whether you're religious or
- 3:19spiritual or atheistic or whatever.
- 3:21Let's say we're living in a simulation.
- 3:22It would make sense
- 3:23>> [music]
- 3:23>> that there is an audit trail for the
- 3:25reality that we're living in. And
- 3:26honestly, I get why this starts feeling
- 3:28so philosophical and spiritual. It's
- 3:30like, what if reality itself kept an
- 3:32audit trail, right? like a record of
- 3:34what happened when and where. Dare I say
- 3:36it's like a browsing history for the
- 3:37real world. I mean, the interesting part
- 3:39is that's also what surveillance is. But
- 3:40a lot of these comments are talking
- 3:41about cosmic surveillance, which is, oh
- 3:43my god, another thing entirely. It's
- 3:45like flock cameras ain't got on the
- 3:47simulation. So, you get the idea. But
- 3:49here, we're recording fundamentally like
- 3:5115 or 20 seconds, and showing you what
- 3:53happens when you visualize it outside of
- 3:55time. But, what happens when you start
- 3:57bringing moments together that are
- 3:59months or even years apart? So, now look
- 4:01at this instead. This is me walking
- 4:03around the same place
- 4:04>> [music]
- 4:05>> in different times simultaneously. Let
- 4:08me explain. So, I 3D scanned this area
- 4:09in the summer, but I've got all these
- 4:11old videos when I recorded it during the
- 4:13winter when it was snowing. What I can
- 4:14do now is figure out the exact location
- 4:16those videos were captured, match it
- 4:19one-to-one to the underlying 3D model,
- 4:21and project the video into 3D space. And
- 4:24look, this is the crazy part to me
- 4:25because we spent so much time going
- 4:27through 2D rectangles in our camera
- 4:29roll. It's a really strange feeling when
- 4:31you can step inside one, almost like
- 4:33navigating through space instead of
- 4:35scrolling through time. I call this the
- 4:37geospatial memory palace because
- 4:38suddenly every place has memory of what
- 4:41was actually observed or recorded there.
- 4:43Now, what happens when you have an event
- 4:45where a lot of people have recorded it
- 4:47simultaneously? For example, TED 2026.
- 4:50So, here I took all the publicly
- 4:52available Creative Commons photos that I
- 4:54could find from TED. Shout out to all
- 4:56the photographers. All the smartphone
- 4:57ones are me, by the way. Now, since a
- 4:59lot of the photos are obviously the red
- 5:01circle, the speaker's on stage, I know
- 5:02exactly the location where all these
- 5:05photos were taken, when they did it, and
- 5:07obviously their main subject of
- 5:09interest. And then I can ask the
- 5:10question, "Hey, how many photos am I
- 5:11in?" Oh, 16 of them. Cool. Show me one
- 5:13of those. And boom, there's me on stage.
- 5:16Now, of course, you know me, 2D isn't
- 5:17enough. So, we can do the same thing and
- 5:19lift every single photo into 3D and
- 5:22leave it exactly where it was taken at
- 5:25the venue. So, you could [music] go
- 5:27backstage. This is the green room.
- 5:30This is Jacob Collier right before he
- 5:32went on stage.
- 5:34This is the CEO of Cloudflare. Shout out
- 5:38Cloudflare. That's Mark Rober on stage.
- 5:41And on and on it goes. Murakami,
- 5:43freaking legit. And there's Steve, the
- 5:45CEO of Reddit during a practice session.
- 5:47So, when you start pulling back and take
- 5:49all these 3D projections and start
- 5:51aggregating them together, it starts
- 5:53feeling a little bit like crowdsourced
- 5:55omniscience. This, to me, is the spatial
- 5:58memory palace taking shape. Every single
- 6:00speaker that was on that stage, exactly
- 6:03where they were, including some members
- 6:04of the audience. But, look, this tech
- 6:06isn't just about revisiting your own
- 6:08memories. It can also help you
- 6:10understand a place that [music] you've
- 6:11never been. So, what you're seeing here
- 6:13is the aftermath of flight 7598, the
- 6:17Prime Air cargo plane that overran the
- 6:19runway in Miami. And the NTSB released
- 6:21drone footage of this location, and I
- 6:23turned it into a 3D [music] model, just
- 6:26like I did the ranch. Once I got the 3D
- 6:28scan, I could even take the actual
- 6:30flight tracking data, shout out God's
- 6:31eye view, and replay it.
- 6:33>> Tower, good afternoon.
- 6:357598 heavy
- 6:37uh ILS 400.
- 6:39>> Cargo South 7598 heavy, wind 300 at 1.
- 6:42>> So, here you see the skid marks of the
- 6:45plane as it overran the runway. This is
- 6:46also where this concept of an audit
- 6:48trail for physical reality becomes
- 6:50useful. There were a bunch of
- 6:51ground-level photos taken. So, here, in
- 6:54this visualization, I can see the
- 6:55impacted vehicle, but it's not super
- 6:58clear. Here's the van, and here's the
- 7:00car. A bunch of people died in this,
- 7:02right? But, I can't really get a sense
- 7:03for what it was like on the ground. So,
- 7:05I just click this button over here, and
- 7:07suddenly all the ground-level photos
- 7:08I've taken
- 7:09>> [music]
- 7:09>> are tightly associated with that 3D
- 7:12model, and that's what you're seeing
- 7:13over here. And by the way, I put a link
- 7:15to this in the description, so you can
- 7:16check out the replay of the flight and
- 7:18jump to any of these moments, too, so
- 7:20you can literally see everything on the
- 7:22ground, as well. Now, this is where it
- 7:24starts connecting back to God's eye
- 7:26view. Putting this reconstruction
- 7:28alongside the flight data, you can see
- 7:30both the third-person and first-person
- 7:32perspective of exactly what went down.
- 7:34You can now go down to the ground and
- 7:36explore the bigger picture. Now, we are
- 7:38not gods, so this This not be the
- 7:40totality of what happened, but it is
- 7:42damn near the closest thing we have to
- 7:44ground truth of what went down. Now, if
- 7:47you pull all the way back out and look
- 7:49at the planet, there's so many
- 7:50satellites that are imaging the planet
- 7:53every single day. I mean, this is the
- 7:55basis of God's eye view. In other words,
- 7:57we have a spatial memory palace for the
- 7:59planet. Check out this visualization I
- 8:01made of these three cyclones going
- 8:03across the Pacific, Lowell, Carina, and
- 8:06Mary. You can literally track them. We
- 8:08can take advantage of all these
- 8:09geostationary satellites and a lot of
- 8:10this data is public. So, here we can
- 8:12zoom in on Lowell and you can see it
- 8:13going from Cat 3 to category 4 to ooh,
- 8:16baby. We can reveal the underlying wind
- 8:19modeling layers of what's going on. You
- 8:21can see its speed right next to it. And
- 8:23here it goes from Cat 3 to just being a
- 8:25little old tropical storm.
- 8:28>> [laughter]
- 8:29>> Now, we don't know for sure where these
- 8:30storms are going to go. So, scientists
- 8:32actually make predictions about what's
- 8:34going to happen forward in time and
- 8:36publish those predictions, which I'm
- 8:37visualizing over here. So, as the storm
- 8:40progresses, you'll see the future path
- 8:42estimates keep changing. So, here's
- 8:44Dujuan's wind visualization and the
- 8:45satellite imagery as it's headed towards
- 8:48Japan over here. So, if you want a sense
- 8:49of where God's eye view is going next,
- 8:52this is a really good hint. I mean, we
- 8:54are building collectively an audit trail
- 8:56for physical reality. We can go back in
- 8:59time and compare this storm to every
- 9:01other storm that's been recorded. You
- 9:02can zoom in on a volcanic eruption from
- 9:05space and see how air traffic is routing
- 9:08around it. Now, we don't have a complete
- 9:10recording of the world. We have pieces
- 9:12and facets of what encompasses an audit
- 9:14trail for physical reality. But like I
- 9:16showed you all with the Iran 4D
- 9:18reconstruction, you could get a
- 9:19remarkably complete picture of what's
- 9:21going down. We started with two deer
- 9:23leaving a 3D trail behind, then a
- 9:25person, a place across seasons, event,
- 9:28and then a planet leaving a record of
- 9:30how it changes. What really pulls me
- 9:32into this is if you take the sensorium,
- 9:34all the the data that we've collected
- 9:36and pull it together into this 4D
- 9:39Commons, this audit trail for physical
- 9:41reality, we get something very profound.
- 9:44Because suddenly it becomes not only a
- 9:45time machine where you can teleport
- 9:47anywhere, it also becomes a bit of a
- 9:49crystal ball giving us a way to figure
- 9:51out what happens next. I'm curious,
- 9:53would you want something like this for
- 9:55your own memories, somewhere where you
- 9:56could upload images and videos and they
- 9:58turn into 3D scans exactly where you
- 10:01left them? Keep the stuff you want
- 10:03private and if there's stuff that you
- 10:05want to open up to the public
- 10:06contributing to this global 4D Commons,
- 10:08this collective audit trail of physical
- 10:10reality, whenever important things
- 10:12happen, we can contribute our data so
- 10:15that the public can make sense of it. I
- 10:17think that would be super sick, but I'm
- 10:18curious if that's something you're into
- 10:20as well. Anyway, hopefully this was fun
- 10:22both below and if you want to see
- 10:26what you can do today, go check out
- 10:28God's Eye View on GitHub. The community
- 10:30has been building some insane new
- 10:31features and there's a lot of updates
- 10:33out there. So if you haven't updated in
- 10:34a minute, go out there and check out
- 10:36what we got cooking. I got a whole new
- 10:38video coming on that. Balavo signing off
- 10:40and I'll see y'all in the next one.
- 10:42Cheers.
About this transcript
This page contains the full transcript of I Built a God’s Eye View of Time. The Internet Went Wild. by Bilawal Sidhu, generated from the public captions YouTube serves with the video. The transcript has 2,190 words across 324 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.