Up next in 10
We were inspired for this podcast episode by the article called:
The Audacious Love Fooled Death (https://doggozila.com/ai-cancer-vaccine-for-dogs/) : AI Cancer Vaccine for Dogs (https://doggozila.com/ai-cancer-vaccine-for-dogs/)
In late 2025, Sydney tech entrepreneur Paul Conyngham received a diagnosis that felt like a digital system failure: his eight-year-old dog, Rose, had aggressive, terminal cancer. Standard veterinary protocols had reached their limit, leaving Rose with a biological death sentence and only months to live. For most, this would be the final chapter, but Conyngham refused to accept the "read-only" nature of the prognosis.
He looked at Rose’s malignancy not as a medical tragedy, but as a complex, solvable data problem. This wasn't about accepting a "shut down" command; it was about refactoring the biological code of the disease itself. By leveraging his expertise in emerging technology, Conyngham launched an algorithmic intervention that would challenge the very boundaries of modern oncology.
Reframing Biology as a Searchable Data Problem
The breakthrough began with a radical mental shift: converting tissue to data. By sequencing the DNA of Rose’s tumor, Conyngham moved the battleground from a physical clinic to a computational environment. Once the genetic information existed as a digital blueprint, the malignancy was no longer a mysterious "black box," but a searchable sequence that could be audited for errors.
This digital-to-biological workflow is the foundational requirement for the next era of personalized medicine. It treats a tumor as a corrupted file within the body's wetware, allowing for precise, targeted edits. For Conyngham, the project was fueled by a deep personal connection that transformed a technical challenge into a mission of love.
Show More Show Less View Video Transcript
0:00
Welcome to another fun episode by the
0:01
Welcome to another fun episode by the
0:01
Welcome to another fun episode by the DogMindGearz podcast.
0:03
DogMindGearz podcast.
0:03
DogMindGearz podcast. >> Glad to be here for this one.
0:04
>> Glad to be here for this one.
0:04
>> Glad to be here for this one. >> Yeah, so I want you to put yourself in a
0:06
>> Yeah, so I want you to put yourself in a
0:06
>> Yeah, so I want you to put yourself in a really difficult position for a second.
0:08
really difficult position for a second.
0:08
really difficult position for a second. >> Okay.
0:08
>> Okay.
0:08
>> Okay. >> Just imagine you're sitting in a sterile
0:10
>> Just imagine you're sitting in a sterile
0:10
>> Just imagine you're sitting in a sterile veterinary clinic.
0:12
veterinary clinic.
0:12
veterinary clinic. Um you hear the hum of the fluorescent
0:15
Um you hear the hum of the fluorescent
0:15
Um you hear the hum of the fluorescent lights buzzing above you.
0:16
lights buzzing above you.
0:16
lights buzzing above you. >> Right, that awful sound.
0:18
>> Right, that awful sound.
0:18
>> Right, that awful sound. >> Exactly. And you've just been handed a
0:19
>> Exactly. And you've just been handed a
0:20
>> Exactly. And you've just been handed a terminal diagnosis for your best friend.
0:22
terminal diagnosis for your best friend.
0:22
terminal diagnosis for your best friend. Your dog.
0:24
Your dog.
0:24
Your dog. What would you do?
0:25
What would you do?
0:25
What would you do? >> I mean, most of us would be completely
0:27
>> I mean, most of us would be completely
0:27
>> I mean, most of us would be completely devastated.
0:28
devastated.
0:28
devastated. >> ask about palliative care, you know,
0:30
>> ask about palliative care, you know,
0:30
>> ask about palliative care, you know, just to keep them comfortable.
0:32
just to keep them comfortable.
0:32
just to keep them comfortable. >> Start preparing to say goodbye.
0:33
>> Start preparing to say goodbye.
0:33
>> Start preparing to say goodbye. >> Right. But today, we are looking at an
0:36
>> Right. But today, we are looking at an
0:36
>> Right. But today, we are looking at an incredible story from Dogazine magazine
0:39
incredible story from Dogazine magazine
0:39
incredible story from Dogazine magazine about a tech entrepreneur who got that
0:42
about a tech entrepreneur who got that
0:42
about a tech entrepreneur who got that exact death sentence for his dog. And
0:45
exact death sentence for his dog. And
0:45
exact death sentence for his dog. And well, he didn't accept it.
0:46
well, he didn't accept it.
0:46
well, he didn't accept it. >> Not at all.
0:47
>> Not at all.
0:47
>> Not at all. >> Yeah, no, instead he used ChatGPT to
0:49
>> Yeah, no, instead he used ChatGPT to
0:49
>> Yeah, no, instead he used ChatGPT to literally hack a cure.
0:51
literally hack a cure.
0:51
literally hack a cure. >> Which sounds like science fiction, but
0:53
>> Which sounds like science fiction, but
0:53
>> Which sounds like science fiction, but the mission for our deep dive today is
0:55
the mission for our deep dive today is
0:55
the mission for our deep dive today is to really break down the mechanics of
0:56
to really break down the mechanics of
0:56
to really break down the mechanics of this. We're going to explore how a
0:58
this. We're going to explore how a
0:58
this. We're going to explore how a biological tragedy was basically
1:00
biological tragedy was basically
1:00
biological tragedy was basically reframed as a data problem.
1:01
reframed as a data problem.
1:01
reframed as a data problem. >> Yeah, pure data problem.
1:03
>> Yeah, pure data problem.
1:03
>> Yeah, pure data problem. >> Exactly. We are going to look at the
1:05
>> Exactly. We are going to look at the
1:05
>> Exactly. We are going to look at the exact pipeline that took raw tumor data,
1:08
exact pipeline that took raw tumor data,
1:08
exact pipeline that took raw tumor data, processed it through artificial
1:09
processed it through artificial
1:09
processed it through artificial intelligence, and uh resulted in a
1:12
intelligence, and uh resulted in a
1:12
intelligence, and uh resulted in a $3,000 custom RNA vaccine.
1:16
$3,000 custom RNA vaccine.
1:16
$3,000 custom RNA vaccine. >> Which is wild.
1:17
>> Which is wild.
1:17
>> Which is wild. >> It is. And more importantly, we need to
1:19
>> It is. And more importantly, we need to
1:19
>> It is. And more importantly, we need to examine what this specific pilot study
1:22
examine what this specific pilot study
1:22
examine what this specific pilot study signals for the future of personalized
1:24
signals for the future of personalized
1:24
signals for the future of personalized medicine.
1:25
medicine.
1:25
medicine. >> For pets and humans, right?
1:26
>> For pets and humans, right?
1:26
>> For pets and humans, right? >> Yeah, not just for the pets sharing our
1:27
>> Yeah, not just for the pets sharing our
1:27
>> Yeah, not just for the pets sharing our homes, but for human medicine down the
1:29
homes, but for human medicine down the
1:29
homes, but for human medicine down the line as well.
1:30
line as well.
1:30
line as well. >> Okay, let's unpack this. Because before
1:32
>> Okay, let's unpack this. Because before
1:32
>> Okay, let's unpack this. Because before we even get into the weeds of artificial
1:34
we even get into the weeds of artificial
1:34
we even get into the weeds of artificial intelligence and RNA synthesis and all
1:36
intelligence and RNA synthesis and all
1:36
intelligence and RNA synthesis and all that, we really need to understand the
1:38
that, we really need to understand the
1:38
that, we really need to understand the massive paradigm shift that made this
1:40
massive paradigm shift that made this
1:40
massive paradigm shift that made this whole thing possible.
1:41
whole thing possible.
1:41
whole thing possible. >> Right, the fundamental change in
1:42
>> Right, the fundamental change in
1:42
>> Right, the fundamental change in perspective.
1:43
perspective.
1:43
perspective. >> Yeah, so this happened in late 2025.
1:46
>> Yeah, so this happened in late 2025.
1:46
>> Yeah, so this happened in late 2025. Paul Cunningham, who is a tech guy
1:48
Paul Cunningham, who is a tech guy
1:48
Paul Cunningham, who is a tech guy living in Sydney, gets the absolute
1:50
living in Sydney, gets the absolute
1:50
living in Sydney, gets the absolute worst news. His 8-year-old dog Rose has
1:53
worst news. His 8-year-old dog Rose has
1:53
worst news. His 8-year-old dog Rose has this aggressive terminal cancer.
1:55
this aggressive terminal cancer.
1:55
this aggressive terminal cancer. >> And she's given months to live.
1:56
>> And she's given months to live.
1:56
>> And she's given months to live. >> Months, Just the ultimate gut punch, you
1:59
>> Months, Just the ultimate gut punch, you
1:59
>> Months, Just the ultimate gut punch, you know.
1:59
know.
1:59
know. >> Oh, absolutely. And it really highlights
2:01
>> Oh, absolutely. And it really highlights
2:01
>> Oh, absolutely. And it really highlights a very specific wall that traditional
2:03
a very specific wall that traditional
2:03
a very specific wall that traditional veterinary medicine often hits.
2:05
veterinary medicine often hits.
2:05
veterinary medicine often hits. >> Yeah, the limits of standard care.
2:06
>> Yeah, the limits of standard care.
2:06
>> Yeah, the limits of standard care. >> Exactly. Because the standard protocols
2:09
>> Exactly. Because the standard protocols
2:09
>> Exactly. Because the standard protocols like chemotherapy and radiation, they
2:11
like chemotherapy and radiation, they
2:11
like chemotherapy and radiation, they are powerful tools, obviously, but they
2:13
are powerful tools, obviously, but they
2:13
are powerful tools, obviously, but they are built on a statistical model.
2:15
are built on a statistical model.
2:15
are built on a statistical model. >> Right.
2:15
>> Right.
2:15
>> Right. >> They're generalized weapons designed for
2:17
>> They're generalized weapons designed for
2:17
>> They're generalized weapons designed for the average patient um based on decades
2:21
the average patient um based on decades
2:21
the average patient um based on decades of broad data.
2:23
of broad data.
2:23
of broad data. >> So, they aren't personalized at all.
2:24
>> So, they aren't personalized at all.
2:24
>> So, they aren't personalized at all. >> No, not at all. When you administer
2:26
>> No, not at all. When you administer
2:26
>> No, not at all. When you administer those treatments, you are basically
2:28
those treatments, you are basically
2:28
those treatments, you are basically flooding the body with systemic
2:29
flooding the body with systemic
2:29
flooding the body with systemic toxicity.
2:31
toxicity.
2:31
toxicity. >> Hoping the cancer dies just a little bit
2:32
>> Hoping the cancer dies just a little bit
2:32
>> Hoping the cancer dies just a little bit faster than the healthy cells do.
2:34
faster than the healthy cells do.
2:34
faster than the healthy cells do. >> Right. It's a race. And for an older dog
2:37
>> Right. It's a race. And for an older dog
2:37
>> Right. It's a race. And for an older dog like Rose, well, her body simply
2:39
like Rose, well, her body simply
2:39
like Rose, well, her body simply couldn't handle that level of collateral
2:41
couldn't handle that level of collateral
2:41
couldn't handle that level of collateral damage.
2:41
damage.
2:41
damage. >> Because traditional oncology in this
2:43
>> Because traditional oncology in this
2:43
>> Because traditional oncology in this context is essentially treating the
2:45
context is essentially treating the
2:45
context is essentially treating the cancer as a uniform biological enemy.
2:48
cancer as a uniform biological enemy.
2:48
cancer as a uniform biological enemy. >> You're throwing a really wide net.
2:49
>> You're throwing a really wide net.
2:49
>> You're throwing a really wide net. >> Exactly.
2:50
>> Exactly.
2:50
>> Exactly. >> Yeah.
2:50
>> Yeah.
2:50
>> Yeah. >> But Cunningham looks at this, and
2:52
>> But Cunningham looks at this, and
2:52
>> But Cunningham looks at this, and because of his tech background, he makes
2:53
because of his tech background, he makes
2:53
because of his tech background, he makes a radical pivot.
2:55
a radical pivot.
2:55
a radical pivot. >> He does.
2:55
>> He does.
2:55
>> He does. >> He decides he isn't going to look at
2:56
>> He decides he isn't going to look at
2:57
>> He decides he isn't going to look at Rose's tumor as this sad, inevitably
2:59
Rose's tumor as this sad, inevitably
2:59
Rose's tumor as this sad, inevitably fail biological specimen. He decides to
3:02
fail biological specimen. He decides to
3:02
fail biological specimen. He decides to look at it as a digital sequence.
3:04
look at it as a digital sequence.
3:04
look at it as a digital sequence. >> And that right there, changing the
3:05
>> And that right there, changing the
3:06
>> And that right there, changing the battleground from the biological realm
3:07
battleground from the biological realm
3:07
battleground from the biological realm to the digital realm, that is what
3:09
to the digital realm, that is what
3:09
to the digital realm, that is what dictates the speed of this entire
3:11
dictates the speed of this entire
3:11
dictates the speed of this entire process.
3:12
process.
3:12
process. >> Because data moves faster than biology.
3:15
>> Because data moves faster than biology.
3:15
>> Because data moves faster than biology. >> Precisely. Cunningham utilized rapid DNA
3:18
>> Precisely. Cunningham utilized rapid DNA
3:18
>> Precisely. Cunningham utilized rapid DNA sequencing. So, instead of just looking
3:20
sequencing. So, instead of just looking
3:20
sequencing. So, instead of just looking at the tissue under a microscope to
3:22
at the tissue under a microscope to
3:22
at the tissue under a microscope to determine the type of cancer
3:23
determine the type of cancer
3:23
determine the type of cancer >> Which is the normal route.
3:24
>> Which is the normal route.
3:24
>> Which is the normal route. >> Right. He took a piece of that physical
3:26
>> Right. He took a piece of that physical
3:26
>> Right. He took a piece of that physical tumor and ran it through a sequencer.
3:28
tumor and ran it through a sequencer.
3:28
tumor and ran it through a sequencer. >> Okay. So, what does that actually do?
3:30
>> Okay. So, what does that actually do?
3:30
>> Okay. So, what does that actually do? >> This machine essentially scans the
3:32
>> This machine essentially scans the
3:32
>> This machine essentially scans the physical matter and outputs gigabytes of
3:35
physical matter and outputs gigabytes of
3:35
physical matter and outputs gigabytes of raw text files.
3:36
raw text files.
3:36
raw text files. >> Wow. Literally turning flesh into text.
3:39
>> Wow. Literally turning flesh into text.
3:39
>> Wow. Literally turning flesh into text. >> Yes.
3:40
>> Yes.
3:40
>> Yes. We are talking about converting a
3:42
We are talking about converting a
3:42
We are talking about converting a chaotic physical malignancy into a
3:45
chaotic physical malignancy into a
3:45
chaotic physical malignancy into a highly readable, organized digital
3:48
highly readable, organized digital
3:48
highly readable, organized digital blueprint.
3:49
blueprint.
3:49
blueprint. >> Containing what? Billions of base pairs
3:51
>> Containing what? Billions of base pairs
3:51
>> Containing what? Billions of base pairs of genetic data.
3:52
of genetic data.
3:52
of genetic data. >> Exactly, billions. And once a tumor
3:54
>> Exactly, billions. And once a tumor
3:54
>> Exactly, billions. And once a tumor exists as a digital data set, you can
3:56
exists as a digital data set, you can
3:56
exists as a digital data set, you can apply massive computational power to it.
3:59
apply massive computational power to it.
3:59
apply massive computational power to it. >> So, it's kind of like you're dealing
4:00
>> So, it's kind of like you're dealing
4:00
>> So, it's kind of like you're dealing with a locked safe, right?
4:01
with a locked safe, right?
4:01
with a locked safe, right? >> that analogy.
4:02
>> that analogy.
4:02
>> that analogy. >> The traditional approach, the
4:03
>> The traditional approach, the
4:03
>> The traditional approach, the chemotherapy approach, is basically
4:05
chemotherapy approach, is basically
4:05
chemotherapy approach, is basically trying to blow the safe up with
4:07
trying to blow the safe up with
4:07
trying to blow the safe up with dynamite.
4:07
dynamite.
4:07
dynamite. >> Right, blunt force.
4:09
>> Right, blunt force.
4:09
>> Right, blunt force. >> Yeah, you might get it open, but you're
4:10
>> Yeah, you might get it open, but you're
4:10
>> Yeah, you might get it open, but you're going to destroy the room in the
4:12
going to destroy the room in the
4:12
going to destroy the room in the process.
4:13
process.
4:13
process. Which in this case is the dog's body.
4:15
Which in this case is the dog's body.
4:15
Which in this case is the dog's body. >> Unfortunately, yes.
4:17
>> Unfortunately, yes.
4:17
>> Unfortunately, yes. >> But by converting the tumor to data,
4:20
>> But by converting the tumor to data,
4:20
>> But by converting the tumor to data, it's like you're sliding a microscopic
4:22
it's like you're sliding a microscopic
4:22
it's like you're sliding a microscopic high-tech scanner inside the safe to map
4:25
high-tech scanner inside the safe to map
4:25
high-tech scanner inside the safe to map the internal mechanics.
4:26
the internal mechanics.
4:26
the internal mechanics. >> To find the exact combination.
4:28
>> To find the exact combination.
4:28
>> To find the exact combination. >> Exactly. And then you unlock it without
4:30
>> Exactly. And then you unlock it without
4:30
>> Exactly. And then you unlock it without breaking a single thing.
4:31
breaking a single thing.
4:31
breaking a single thing. >> That's the ideal scenario, yes.
4:33
>> That's the ideal scenario, yes.
4:33
>> That's the ideal scenario, yes. >> But here is the thing that really trips
4:34
>> But here is the thing that really trips
4:34
>> But here is the thing that really trips me up. You scan the safe, right? You get
4:37
me up. You scan the safe, right? You get
4:37
me up. You scan the safe, right? You get the blueprint.
4:38
the blueprint.
4:38
the blueprint. >> Right.
4:38
>> Right.
4:38
>> Right. >> It's just a massive data set of genetic
4:41
>> It's just a massive data set of genetic
4:41
>> It's just a massive data set of genetic information.
4:42
information.
4:42
information. How does a tech guy read a biological
4:45
How does a tech guy read a biological
4:45
How does a tech guy read a biological blueprint of that magnitude?
4:47
blueprint of that magnitude?
4:47
blueprint of that magnitude? >> Well, the short answer is he doesn't.
4:49
>> Well, the short answer is he doesn't.
4:49
>> Well, the short answer is he doesn't. >> Oh.
4:49
>> Oh.
4:49
>> Oh. >> When dealing with genomic data sets,
4:51
>> When dealing with genomic data sets,
4:51
>> When dealing with genomic data sets, manual analysis is computationally
4:53
manual analysis is computationally
4:53
manual analysis is computationally impossible for a human brain.
4:55
impossible for a human brain.
4:55
impossible for a human brain. >> Because of the size?
4:56
>> Because of the size?
4:56
>> Because of the size? >> Yeah. A single tumor genome contains
4:59
>> Yeah. A single tumor genome contains
4:59
>> Yeah. A single tumor genome contains billions of letters of genetic code. And
5:02
billions of letters of genetic code. And
5:02
billions of letters of genetic code. And the mutations driving the cancer are
5:03
the mutations driving the cancer are
5:03
the mutations driving the cancer are often just a handful of misspellings
5:05
often just a handful of misspellings
5:05
often just a handful of misspellings hidden within that massive library.
5:07
hidden within that massive library.
5:07
hidden within that massive library. >> Like finding a needle in a digital
5:08
>> Like finding a needle in a digital
5:08
>> Like finding a needle in a digital haystack.
5:09
haystack.
5:09
haystack. >> Worse than a haystack, honestly.
5:11
>> Worse than a haystack, honestly.
5:11
>> Worse than a haystack, honestly. >> Yeah.
5:11
>> Yeah.
5:11
>> Yeah. >> But we have to view DNA fundamentally as
5:13
>> But we have to view DNA fundamentally as
5:13
>> But we have to view DNA fundamentally as a language.
5:14
a language.
5:14
a language. >> Okay, a language like English or
5:15
>> Okay, a language like English or
5:15
>> Okay, a language like English or Spanish?
5:16
Spanish?
5:16
Spanish? >> Sort of. It has syntax, it has grammar,
5:19
>> Sort of. It has syntax, it has grammar,
5:19
>> Sort of. It has syntax, it has grammar, and it has structural rules that dictate
5:21
and it has structural rules that dictate
5:21
and it has structural rules that dictate how a biological organism operates.
5:23
how a biological organism operates.
5:23
how a biological organism operates. You've got the letters A, C, G, and T.
5:26
You've got the letters A, C, G, and T.
5:26
You've got the letters A, C, G, and T. >> The basic building blocks.
5:28
>> The basic building blocks.
5:28
>> The basic building blocks. >> Exactly. When cancer occurs, it is a
5:30
>> Exactly. When cancer occurs, it is a
5:30
>> Exactly. When cancer occurs, it is a grammatical error in that code telling a
5:33
grammatical error in that code telling a
5:33
grammatical error in that code telling a cell to multiply uncontrollably.
5:35
cell to multiply uncontrollably.
5:35
cell to multiply uncontrollably. >> A glitch in the software.
5:36
>> A glitch in the software.
5:36
>> A glitch in the software. >> Right. So, finding that error requires a
5:39
>> Right. So, finding that error requires a
5:39
>> Right. So, finding that error requires a tool designed to parse massive amounts
5:41
tool designed to parse massive amounts
5:41
tool designed to parse massive amounts of language-based data at lightning
5:43
of language-based data at lightning
5:43
of language-based data at lightning speed.
5:44
speed.
5:44
speed. >> And this is where the story takes a wild
5:45
>> And this is where the story takes a wild
5:45
>> And this is where the story takes a wild turn.
5:46
turn.
5:46
turn. >> It really does.
5:47
>> It really does.
5:47
>> It really does. >> Because to find that type O, Cunningham
5:49
>> Because to find that type O, Cunningham
5:49
>> Because to find that type O, Cunningham turns to chat GPT.
5:50
turns to chat GPT.
5:50
turns to chat GPT. >> Yep, off-the-shelf AI.
5:52
>> Yep, off-the-shelf AI.
5:52
>> Yep, off-the-shelf AI. >> He takes this raw genetic data set,
5:54
>> He takes this raw genetic data set,
5:54
>> He takes this raw genetic data set, plugs it into the AI, and uses it as a
5:57
plugs it into the AI, and uses it as a
5:57
plugs it into the AI, and uses it as a high-speed linguistic debugger for his
5:59
high-speed linguistic debugger for his
5:59
high-speed linguistic debugger for his dog's DNA.
6:00
dog's DNA.
6:00
dog's DNA. >> Which is brilliant when you think about
6:01
>> Which is brilliant when you think about
6:01
>> Which is brilliant when you think about it.
6:02
it.
6:02
it. >> It's insane. He's asking it to scan the
6:03
>> It's insane. He's asking it to scan the
6:04
>> It's insane. He's asking it to scan the genomic landscape and pinpoint the
6:06
genomic landscape and pinpoint the
6:06
genomic landscape and pinpoint the unique mutations, the specific glitches
6:08
unique mutations, the specific glitches
6:08
unique mutations, the specific glitches that are driving Rose's individual
6:10
that are driving Rose's individual
6:10
that are driving Rose's individual cancer.
6:11
cancer.
6:11
cancer. >> And large language models like chat GPT
6:13
>> And large language models like chat GPT
6:13
>> And large language models like chat GPT are at their very foundation advanced
6:15
are at their very foundation advanced
6:15
are at their very foundation advanced pattern recognition engine.
6:16
pattern recognition engine.
6:17
pattern recognition engine. >> Right, they predict the next word.
6:18
>> Right, they predict the next word.
6:18
>> Right, they predict the next word. >> Yeah, they're trained to predict and
6:19
>> Yeah, they're trained to predict and
6:19
>> Yeah, they're trained to predict and identify linguistic patterns based on
6:21
identify linguistic patterns based on
6:21
identify linguistic patterns based on context. While we normally use them for
6:24
context. While we normally use them for
6:24
context. While we normally use them for English or, you know, Python code,
6:26
English or, you know, Python code,
6:26
English or, you know, Python code, genetic sequencing is just another
6:28
genetic sequencing is just another
6:28
genetic sequencing is just another syntax.
6:29
syntax.
6:29
syntax. >> It's just letters in a row.
6:30
>> It's just letters in a row.
6:30
>> It's just letters in a row. >> Exactly. So, when you feed it tumor data
6:33
>> Exactly. So, when you feed it tumor data
6:33
>> Exactly. So, when you feed it tumor data alongside healthy baseline data, the AI
6:36
alongside healthy baseline data, the AI
6:36
alongside healthy baseline data, the AI applies its pattern matching
6:38
applies its pattern matching
6:38
applies its pattern matching capabilities to filter out the noise.
6:40
capabilities to filter out the noise.
6:40
capabilities to filter out the noise. >> It compares the two files, basically.
6:42
>> It compares the two files, basically.
6:42
>> It compares the two files, basically. >> Right, it isolates the anomalies, the
6:45
>> Right, it isolates the anomalies, the
6:45
>> Right, it isolates the anomalies, the deviations from the healthy grammar, if
6:47
deviations from the healthy grammar, if
6:47
deviations from the healthy grammar, if you will.
6:48
you will.
6:48
you will. And it flags those specific genetic
6:50
And it flags those specific genetic
6:50
And it flags those specific genetic mutations as the drivers of the
6:52
mutations as the drivers of the
6:52
mutations as the drivers of the malignancy.
6:53
malignancy.
6:53
malignancy. >> Okay, wait, I have to play the skeptic
6:54
>> Okay, wait, I have to play the skeptic
6:55
>> Okay, wait, I have to play the skeptic here for a second.
6:55
here for a second.
6:55
here for a second. >> Go for it.
6:56
>> Go for it.
6:56
>> Go for it. >> Chat GPT is trained on internet text,
6:58
>> Chat GPT is trained on internet text,
6:58
>> Chat GPT is trained on internet text, right? It aggregates Reddit threads,
7:00
right? It aggregates Reddit threads,
7:00
right? It aggregates Reddit threads, Wikipedia, books.
7:01
Wikipedia, books.
7:01
Wikipedia, books. >> Yes, massive amounts of web text.
7:02
>> Yes, massive amounts of web text.
7:03
>> Yes, massive amounts of web text. >> It doesn't have a medical degree, and it
7:04
>> It doesn't have a medical degree, and it
7:04
>> It doesn't have a medical degree, and it certainly isn't a veterinary oncologist.
7:06
certainly isn't a veterinary oncologist.
7:06
certainly isn't a veterinary oncologist. >> True.
7:07
>> True.
7:07
>> True. >> How does it know how to process the
7:08
>> How does it know how to process the
7:08
>> How does it know how to process the syntax of genetic sequencing without
7:11
syntax of genetic sequencing without
7:11
syntax of genetic sequencing without just hallucinating or making things up
7:13
just hallucinating or making things up
7:13
just hallucinating or making things up entirely? I mean, we've all seen AI
7:16
entirely? I mean, we've all seen AI
7:16
entirely? I mean, we've all seen AI confidently give completely wrong, even
7:19
confidently give completely wrong, even
7:19
confidently give completely wrong, even unhinged answers.
7:20
unhinged answers.
7:20
unhinged answers. >> This raises an important question, and
7:22
>> This raises an important question, and
7:22
>> This raises an important question, and it's the exact reason why this project
7:24
it's the exact reason why this project
7:24
it's the exact reason why this project couldn't just happen in a guy's garage.
7:26
couldn't just happen in a guy's garage.
7:26
couldn't just happen in a guy's garage. >> Right, because of the risk.
7:27
>> Right, because of the risk.
7:27
>> Right, because of the risk. >> Exactly.
7:28
>> Exactly.
7:28
>> Exactly. If an AI hallucinates a genetic target,
7:31
If an AI hallucinates a genetic target,
7:31
If an AI hallucinates a genetic target, meaning it flags a healthy, normal
7:33
meaning it flags a healthy, normal
7:33
meaning it flags a healthy, normal genetic sequence as a cancer mutation,
7:36
genetic sequence as a cancer mutation,
7:36
genetic sequence as a cancer mutation, the results are catastrophic.
7:37
the results are catastrophic.
7:37
the results are catastrophic. >> What happens then?
7:39
>> What happens then?
7:39
>> What happens then? >> Well, if you build a vaccine based on
7:41
>> Well, if you build a vaccine based on
7:41
>> Well, if you build a vaccine based on that false target, you will train the
7:43
that false target, you will train the
7:43
that false target, you will train the dog's immune system to attack its own
7:45
dog's immune system to attack its own
7:45
dog's immune system to attack its own healthy tissue.
7:46
healthy tissue.
7:46
healthy tissue. >> Oh, wow.
7:47
>> Oh, wow.
7:47
>> Oh, wow. >> Yeah, you would induce a severe,
7:49
>> Yeah, you would induce a severe,
7:49
>> Yeah, you would induce a severe, potentially lethal autoimmune disease.
7:51
potentially lethal autoimmune disease.
7:51
potentially lethal autoimmune disease. >> That's terrifying.
7:52
>> That's terrifying.
7:52
>> That's terrifying. >> It is. The AI is brilliant at filtering
7:55
>> It is. The AI is brilliant at filtering
7:55
>> It is. The AI is brilliant at filtering the data set and spotting anomalies
7:57
the data set and spotting anomalies
7:57
the data set and spotting anomalies quickly, but that digital insight is
7:59
quickly, but that digital insight is
7:59
quickly, but that digital insight is merely a hypothesis.
8:01
merely a hypothesis.
8:01
merely a hypothesis. >> So, it's not the final answer.
8:02
>> So, it's not the final answer.
8:02
>> So, it's not the final answer. >> Not at all. It must be subjected to
8:04
>> Not at all. It must be subjected to
8:04
>> Not at all. It must be subjected to rigorous human scientific validation.
8:07
rigorous human scientific validation.
8:07
rigorous human scientific validation. >> Okay, so the AI prompt isn't the cure
8:09
>> Okay, so the AI prompt isn't the cure
8:09
>> Okay, so the AI prompt isn't the cure itself. It's just a heavily educated
8:11
itself. It's just a heavily educated
8:11
itself. It's just a heavily educated guess that needs to be fact-checked by
8:13
guess that needs to be fact-checked by
8:13
guess that needs to be fact-checked by real scientists.
8:14
real scientists.
8:14
real scientists. >> Precisely. And the next phase of this
8:16
>> Precisely. And the next phase of this
8:16
>> Precisely. And the next phase of this story, the academic collaboration, is
8:18
story, the academic collaboration, is
8:18
story, the academic collaboration, is where those safety mechanisms are
8:20
where those safety mechanisms are
8:20
where those safety mechanisms are actually applied.
8:21
actually applied.
8:21
actually applied. >> Right, because he didn't do this alone.
8:22
>> Right, because he didn't do this alone.
8:22
>> Right, because he didn't do this alone. >> No, Cunningham partnered with Professor
8:24
>> No, Cunningham partnered with Professor
8:24
>> No, Cunningham partnered with Professor Pall Thordarson and the UNSW RNA
8:26
Pall Thordarson and the UNSW RNA
8:27
Pall Thordarson and the UNSW RNA Institute over in Australia.
8:28
Institute over in Australia.
8:28
Institute over in Australia. >> Which is a major research hub.
8:30
>> Which is a major research hub.
8:30
>> Which is a major research hub. >> Huge.
8:31
>> Huge.
8:31
>> Huge. And the scientists at the university
8:33
And the scientists at the university
8:33
And the scientists at the university took the mutations that ChatGPT flagged
8:36
took the mutations that ChatGPT flagged
8:36
took the mutations that ChatGPT flagged and ran them against known, validated
8:38
and ran them against known, validated
8:38
and ran them against known, validated biological databases.
8:40
biological databases.
8:40
biological databases. >> Like checking the AI's homework.
8:41
>> Like checking the AI's homework.
8:41
>> Like checking the AI's homework. >> Exactly like that. They used things like
8:43
>> Exactly like that. They used things like
8:43
>> Exactly like that. They used things like the canine genome project. They had to
8:46
the canine genome project. They had to
8:46
the canine genome project. They had to manually ensure that the genetic targets
8:48
manually ensure that the genetic targets
8:48
manually ensure that the genetic targets the AI found were actually oncogenes.
8:50
the AI found were actually oncogenes.
8:50
the AI found were actually oncogenes. >> Meaning known cancer drivers.
8:52
>> Meaning known cancer drivers.
8:52
>> Meaning known cancer drivers. >> Right. And that they were biologically
8:54
>> Right. And that they were biologically
8:54
>> Right. And that they were biologically actionable. They acted as this vital
8:57
actionable. They acted as this vital
8:57
actionable. They acted as this vital institutional safety net, translating
8:59
institutional safety net, translating
8:59
institutional safety net, translating the AI's rapid data synthesis into
9:01
the AI's rapid data synthesis into
9:02
the AI's rapid data synthesis into scientifically sound medical targets.
9:04
scientifically sound medical targets.
9:04
scientifically sound medical targets. >> That dynamic is just incredible to me.
9:06
>> That dynamic is just incredible to me.
9:06
>> That dynamic is just incredible to me. It's the ultimate team-up between a
9:08
It's the ultimate team-up between a
9:08
It's the ultimate team-up between a citizen scientist and an academic
9:10
citizen scientist and an academic
9:10
citizen scientist and an academic institution.
9:11
institution.
9:11
institution. >> It's a very modern partnership.
9:13
>> It's a very modern partnership.
9:13
>> It's a very modern partnership. >> tech entrepreneur brings this insane
9:14
>> tech entrepreneur brings this insane
9:14
>> tech entrepreneur brings this insane speed and a high-risk tolerance, and the
9:17
speed and a high-risk tolerance, and the
9:17
speed and a high-risk tolerance, and the university brings the rigor, the safety
9:19
university brings the rigor, the safety
9:20
university brings the rigor, the safety protocols, and the literal laboratory
9:22
protocols, and the literal laboratory
9:22
protocols, and the literal laboratory infrastructure to error-check the AI.
9:24
infrastructure to error-check the AI.
9:24
infrastructure to error-check the AI. >> Because without that error-checking, you
9:26
>> Because without that error-checking, you
9:26
>> Because without that error-checking, you have nothing safe to use.
9:27
have nothing safe to use.
9:27
have nothing safe to use. >> Right. And once they validated the data
9:30
>> Right. And once they validated the data
9:30
>> Right. And once they validated the data and they confirmed the AI didn't
9:31
and they confirmed the AI didn't
9:31
and they confirmed the AI didn't hallucinate, they had actually build the
9:33
hallucinate, they had actually build the
9:33
hallucinate, they had actually build the thing.
9:34
thing.
9:34
thing. >> The physical vaccine.
9:35
>> The physical vaccine.
9:35
>> The physical vaccine. >> Yeah. Can you break down that pipeline?
9:37
>> Yeah. Can you break down that pipeline?
9:37
>> Yeah. Can you break down that pipeline? Like, how do you physically go from a
9:38
Like, how do you physically go from a
9:38
Like, how do you physically go from a verified digital text file to a vial of
9:42
verified digital text file to a vial of
9:42
verified digital text file to a vial of medicine?
9:43
medicine?
9:43
medicine? >> So, the pipeline relies on synthetic
9:46
>> So, the pipeline relies on synthetic
9:46
>> So, the pipeline relies on synthetic messenger RNA, or mRNA.
9:48
messenger RNA, or mRNA.
9:48
messenger RNA, or mRNA. >> Which we're somewhat familiar with now.
9:49
>> Which we're somewhat familiar with now.
9:50
>> Which we're somewhat familiar with now. >> Yes, thanks to recent years. Once the
9:52
>> Yes, thanks to recent years. Once the
9:52
>> Yes, thanks to recent years. Once the exact genetic sequence of the tumor's
9:54
exact genetic sequence of the tumor's
9:54
exact genetic sequence of the tumor's mutation is confirmed on a computer
9:56
mutation is confirmed on a computer
9:56
mutation is confirmed on a computer screen, the lab uses specialized
9:58
screen, the lab uses specialized
9:58
screen, the lab uses specialized equipment in a process called in vitro
10:00
equipment in a process called in vitro
10:00
equipment in a process called in vitro transcription.
10:01
transcription.
10:01
transcription. >> Sounds complicated.
10:02
>> Sounds complicated.
10:02
>> Sounds complicated. >> It basically synthesizes a physical mRNA
10:05
>> It basically synthesizes a physical mRNA
10:05
>> It basically synthesizes a physical mRNA strand that perfectly mirrors that
10:07
strand that perfectly mirrors that
10:07
strand that perfectly mirrors that cancer mutation.
10:08
cancer mutation.
10:08
cancer mutation. >> Okay, so they print the instructions.
10:10
>> Okay, so they print the instructions.
10:10
>> Okay, so they print the instructions. >> Right. But, you can't just inject raw
10:12
>> Right. But, you can't just inject raw
10:12
>> Right. But, you can't just inject raw mRNA into a dog's arm.
10:14
mRNA into a dog's arm.
10:14
mRNA into a dog's arm. >> Why not?
10:15
>> Why not?
10:15
>> Why not? >> Well, enzymes in the bloodstream would
10:16
>> Well, enzymes in the bloodstream would
10:16
>> Well, enzymes in the bloodstream would shred it instantly. It's very fragile.
10:18
shred it instantly. It's very fragile.
10:18
shred it instantly. It's very fragile. It has to be packaged.
10:20
It has to be packaged.
10:20
It has to be packaged. >> Ah, okay. How do they package it?
10:22
>> Ah, okay. How do they package it?
10:22
>> Ah, okay. How do they package it? >> The lab encapsulates this delicate mRNA
10:24
>> The lab encapsulates this delicate mRNA
10:24
>> The lab encapsulates this delicate mRNA inside lipid nanoparticles.
10:26
inside lipid nanoparticles.
10:27
inside lipid nanoparticles. >> Lipid nanoparticles.
10:28
>> Lipid nanoparticles.
10:28
>> Lipid nanoparticles. >> Yeah, which are essentially microscopic
10:30
>> Yeah, which are essentially microscopic
10:30
>> Yeah, which are essentially microscopic fat bubbles.
10:31
fat bubbles.
10:31
fat bubbles. >> Oh, fat bubbles. That makes sense.
10:32
>> Oh, fat bubbles. That makes sense.
10:32
>> Oh, fat bubbles. That makes sense. >> These bubbles protect the mRNA carrier
10:35
>> These bubbles protect the mRNA carrier
10:35
>> These bubbles protect the mRNA carrier as it travels through the bloodstream,
10:36
as it travels through the bloodstream,
10:36
as it travels through the bloodstream, allowing it to slip safely inside the
10:38
allowing it to slip safely inside the
10:38
allowing it to slip safely inside the dog's immune cells.
10:39
dog's immune cells.
10:39
dog's immune cells. >> And once it's inside?
10:41
>> And once it's inside?
10:41
>> And once it's inside? >> Once inside, the mRNA delivers the
10:43
>> Once inside, the mRNA delivers the
10:43
>> Once inside, the mRNA delivers the instructions, essentially handing the
10:45
instructions, essentially handing the
10:45
instructions, essentially handing the body's natural defenses a highly
10:48
body's natural defenses a highly
10:48
body's natural defenses a highly detailed wanted poster of the cancer's
10:51
detailed wanted poster of the cancer's
10:51
detailed wanted poster of the cancer's unique genetic signature.
10:52
unique genetic signature.
10:52
unique genetic signature. >> A wanted poster. I love that. And the
10:55
>> A wanted poster. I love that. And the
10:55
>> A wanted poster. I love that. And the most mind-blowing part of this whole
10:57
most mind-blowing part of this whole
10:57
most mind-blowing part of this whole pipeline is the cost and the speed.
10:59
pipeline is the cost and the speed.
10:59
pipeline is the cost and the speed. >> It really is unprecedented.
11:01
>> It really is unprecedented.
11:01
>> It really is unprecedented. >> Right. Because we are so conditioned to
11:03
>> Right. Because we are so conditioned to
11:03
>> Right. Because we are so conditioned to expect medical breakthroughs to take a
11:05
expect medical breakthroughs to take a
11:05
expect medical breakthroughs to take a decade of trials and billions of dollars
11:07
decade of trials and billions of dollars
11:07
decade of trials and billions of dollars in funding.
11:08
in funding.
11:08
in funding. >> That's the traditional model.
11:09
>> That's the traditional model.
11:09
>> That's the traditional model. >> Yeah. And Professor Thordarson himself
11:12
>> Yeah. And Professor Thordarson himself
11:12
>> Yeah. And Professor Thordarson himself admitted he was incredibly skeptical
11:14
admitted he was incredibly skeptical
11:14
admitted he was incredibly skeptical when Cunningham first approached him.
11:15
when Cunningham first approached him.
11:15
when Cunningham first approached him. >> Of course, he was. He's a traditional
11:17
>> Of course, he was. He's a traditional
11:17
>> Of course, he was. He's a traditional academic.
11:18
academic.
11:18
academic. >> Right, he didn't think an academic lab
11:19
>> Right, he didn't think an academic lab
11:19
>> Right, he didn't think an academic lab could move fast enough to save Rose
11:21
could move fast enough to save Rose
11:21
could move fast enough to save Rose before the cancer took her.
11:22
before the cancer took her.
11:22
before the cancer took her. >> But, they did.
11:23
>> But, they did.
11:23
>> But, they did. >> They did. From the initial sequencing
11:25
>> They did. From the initial sequencing
11:25
>> They did. From the initial sequencing through the AI debugging all the way to
11:27
through the AI debugging all the way to
11:27
through the AI debugging all the way to a manufactured sterile
11:29
a manufactured sterile
11:29
a manufactured sterile lipid-encapsulated RNA vaccine.
11:31
lipid-encapsulated RNA vaccine.
11:32
lipid-encapsulated RNA vaccine. >> The whole thing.
11:32
>> The whole thing.
11:32
>> The whole thing. >> It took less than 2 months.
11:33
>> It took less than 2 months.
11:33
>> It took less than 2 months. >> Unbelievable.
11:34
>> Unbelievable.
11:34
>> Unbelievable. >> And the out-of-pocket cost was
11:35
>> And the out-of-pocket cost was
11:36
>> And the out-of-pocket cost was approximately $3,000.
11:38
approximately $3,000.
11:38
approximately $3,000. >> And that speed fundamentally disrupts
11:40
>> And that speed fundamentally disrupts
11:40
>> And that speed fundamentally disrupts the current medical paradigm. By
11:42
the current medical paradigm. By
11:42
the current medical paradigm. By treating the cancer as an information
11:44
treating the cancer as an information
11:44
treating the cancer as an information problem first, they completely bypass
11:47
problem first, they completely bypass
11:47
problem first, they completely bypass the years required for broad generalized
11:49
the years required for broad generalized
11:49
the years required for broad generalized clinical trials.
11:50
clinical trials.
11:50
clinical trials. >> They skipped the line entirely.
11:52
>> They skipped the line entirely.
11:52
>> They skipped the line entirely. >> Yeah, they went straight from data
11:53
>> Yeah, they went straight from data
11:53
>> Yeah, they went straight from data analysis to a custom-manufactured
11:56
analysis to a custom-manufactured
11:56
analysis to a custom-manufactured biological solution tailored for a
11:58
biological solution tailored for a
11:58
biological solution tailored for a sample size of exactly one.
11:59
sample size of exactly one.
12:00
sample size of exactly one. >> Sample size of one. That's the core of
12:01
>> Sample size of one. That's the core of
12:01
>> Sample size of one. That's the core of personalized medicine right there.
12:03
personalized medicine right there.
12:03
personalized medicine right there. >> Absolutely.
12:03
>> Absolutely.
12:03
>> Absolutely. >> Okay, so they have the physical vaccine
12:05
>> Okay, so they have the physical vaccine
12:05
>> Okay, so they have the physical vaccine they administer to Rose, but
12:06
they administer to Rose, but
12:06
they administer to Rose, but immunotherapy isn't like movie magic
12:08
immunotherapy isn't like movie magic
12:08
immunotherapy isn't like movie magic where the patient gets a shot and
12:09
where the patient gets a shot and
12:09
where the patient gets a shot and instantly leaps out of bed, right?
12:11
instantly leaps out of bed, right?
12:11
instantly leaps out of bed, right? >> No, not at all.
12:12
>> No, not at all.
12:12
>> No, not at all. >> There is a real-world biological and
12:14
>> There is a real-world biological and
12:14
>> There is a real-world biological and emotional timeline here that requires a
12:16
emotional timeline here that requires a
12:16
emotional timeline here that requires a massive amount of patience.
12:18
massive amount of patience.
12:18
massive amount of patience. >> Yeah, the waiting game is notoriously
12:20
>> Yeah, the waiting game is notoriously
12:20
>> Yeah, the waiting game is notoriously difficult with immunotherapy.
12:22
difficult with immunotherapy.
12:22
difficult with immunotherapy. >> Because it's not instantaneous.
12:23
>> Because it's not instantaneous.
12:23
>> Because it's not instantaneous. >> Right. Unlike a painkiller that binds to
12:26
>> Right. Unlike a painkiller that binds to
12:26
>> Right. Unlike a painkiller that binds to a receptor and offers immediate relief,
12:29
a receptor and offers immediate relief,
12:29
a receptor and offers immediate relief, an RNA vaccine requires a biological lag
12:32
an RNA vaccine requires a biological lag
12:32
an RNA vaccine requires a biological lag time.
12:32
time.
12:32
time. >> A lag time. How long are we talking?
12:34
>> A lag time. How long are we talking?
12:34
>> A lag time. How long are we talking? >> Well, when the lipid nanoparticles
12:36
>> Well, when the lipid nanoparticles
12:37
>> Well, when the lipid nanoparticles deliver that blueprint into the immune
12:38
deliver that blueprint into the immune
12:38
deliver that blueprint into the immune cells, you aren't directly killing the
12:41
cells, you aren't directly killing the
12:41
cells, you aren't directly killing the cancer right then and there.
12:42
cancer right then and there.
12:42
cancer right then and there. >> Oh, right. You're just delivering the
12:43
>> Oh, right. You're just delivering the
12:43
>> Oh, right. You're just delivering the wanted poster.
12:44
wanted poster.
12:45
wanted poster. >> Exactly. You are running a boot camp for
12:46
>> Exactly. You are running a boot camp for
12:46
>> Exactly. You are running a boot camp for the immune system.
12:48
the immune system.
12:48
the immune system. The specialized immune cells,
12:50
The specialized immune cells,
12:50
The specialized immune cells, specifically dendritic cells, have to
12:52
specifically dendritic cells, have to
12:52
specifically dendritic cells, have to translate the mRNA.
12:53
translate the mRNA.
12:53
translate the mRNA. >> Okay.
12:54
>> Okay.
12:54
>> Okay. >> Then they display the cancer antigen on
12:56
>> Then they display the cancer antigen on
12:56
>> Then they display the cancer antigen on their surface, and they travel to the
12:58
their surface, and they travel to the
12:58
their surface, and they travel to the lymph nodes to train the T cells.
13:00
lymph nodes to train the T cells.
13:00
lymph nodes to train the T cells. >> That sounds like a lot of steps.
13:01
>> That sounds like a lot of steps.
13:01
>> That sounds like a lot of steps. >> It is. Those T cells then have to
13:04
>> It is. Those T cells then have to
13:04
>> It is. Those T cells then have to multiply into an army large enough to
13:05
multiply into an army large enough to
13:06
multiply into an army large enough to actually mount a systemic defense.
13:08
actually mount a systemic defense.
13:08
actually mount a systemic defense. >> And that takes time.
13:08
>> And that takes time.
13:08
>> And that takes time. >> Yeah, that cellular machinery takes
13:10
>> Yeah, that cellular machinery takes
13:10
>> Yeah, that cellular machinery takes weeks to fully activate.
13:12
weeks to fully activate.
13:12
weeks to fully activate. >> And you can really see that agonizing
13:14
>> And you can really see that agonizing
13:14
>> And you can really see that agonizing lag in Rose's timeline.
13:15
lag in Rose's timeline.
13:16
lag in Rose's timeline. >> It must have been terrifying for the
13:17
>> It must have been terrifying for the
13:17
>> It must have been terrifying for the owner.
13:17
owner.
13:17
owner. >> Totally.
13:18
>> Totally.
13:18
>> Totally. In early December, right around the time
13:20
In early December, right around the time
13:21
In early December, right around the time of the diagnosis, Cunningham describes
13:23
of the diagnosis, Cunningham describes
13:23
of the diagnosis, Cunningham describes her as immobile, sad, and completely
13:26
her as immobile, sad, and completely
13:26
her as immobile, sad, and completely shut down.
13:27
shut down.
13:27
shut down. >> Just a sick, sick dog.
13:29
>> Just a sick, sick dog.
13:29
>> Just a sick, sick dog. >> Yeah. They administer the vaccine and
13:30
>> Yeah. They administer the vaccine and
13:30
>> Yeah. They administer the vaccine and for weeks it probably feels like
13:32
for weeks it probably feels like
13:32
for weeks it probably feels like absolutely nothing is happening on the
13:34
absolutely nothing is happening on the
13:34
absolutely nothing is happening on the outside.
13:34
outside.
13:34
outside. >> But beneath the surface, her immune
13:36
>> But beneath the surface, her immune
13:36
>> But beneath the surface, her immune system is going through that exact boot
13:38
system is going through that exact boot
13:38
system is going through that exact boot camp you described.
13:39
camp you described.
13:39
camp you described. >> Right. And then we hit late January, so
13:41
>> Right. And then we hit late January, so
13:41
>> Right. And then we hit late January, so about a month or so later, and the
13:43
about a month or so later, and the
13:43
about a month or so later, and the physical recovery changes are just
13:45
physical recovery changes are just
13:45
physical recovery changes are just staggering.
13:46
staggering.
13:46
staggering. >> It's a night and day difference.
13:47
>> It's a night and day difference.
13:47
>> It's a night and day difference. >> Rose goes from being completely immobile
13:49
>> Rose goes from being completely immobile
13:49
>> Rose goes from being completely immobile to literally jumping over fences to
13:51
to literally jumping over fences to
13:51
to literally jumping over fences to chase rabbits in the yard.
13:53
chase rabbits in the yard.
13:53
chase rabbits in the yard. >> Incredible.
13:54
>> Incredible.
13:54
>> Incredible. >> Her energy returns entirely.
13:56
>> Her energy returns entirely.
13:56
>> Her energy returns entirely. And when they finally ran the scans,
13:57
And when they finally ran the scans,
13:57
And when they finally ran the scans, statistically, they achieved a 75%
14:00
statistically, they achieved a 75%
14:00
statistically, they achieved a 75% reduction in the cancer.
14:01
reduction in the cancer.
14:01
reduction in the cancer. >> And we really need to pause on that
14:02
>> And we really need to pause on that
14:02
>> And we really need to pause on that number. In the context of a terminal,
14:06
number. In the context of a terminal,
14:06
number. In the context of a terminal, aggressive oncology diagnosis, reducing
14:09
aggressive oncology diagnosis, reducing
14:09
aggressive oncology diagnosis, reducing a tumor mass by 3/4 is an absolute
14:12
a tumor mass by 3/4 is an absolute
14:12
a tumor mass by 3/4 is an absolute clinical triumph.
14:14
clinical triumph.
14:14
clinical triumph. >> It's massive.
14:15
>> It's massive.
14:15
>> It's massive. >> It effectively shifts the patient's
14:16
>> It effectively shifts the patient's
14:16
>> It effectively shifts the patient's status from immediate end-of-life to
14:19
status from immediate end-of-life to
14:19
status from immediate end-of-life to managed chronic care.
14:20
managed chronic care.
14:20
managed chronic care. >> Buying them so much time.
14:22
>> Buying them so much time.
14:22
>> Buying them so much time. >> Yes. It stabilizes the biological
14:24
>> Yes. It stabilizes the biological
14:24
>> Yes. It stabilizes the biological environment, relieves the pressure on
14:26
environment, relieves the pressure on
14:26
environment, relieves the pressure on surrounding organs, and removes the
14:28
surrounding organs, and removes the
14:28
surrounding organs, and removes the immediate lethal threat.
14:30
immediate lethal threat.
14:30
immediate lethal threat. >> But here's where it gets really
14:31
>> But here's where it gets really
14:31
>> But here's where it gets really interesting, though.
14:32
interesting, though.
14:32
interesting, though. And I have to push back a little on this
14:34
And I have to push back a little on this
14:34
And I have to push back a little on this being a total victory.
14:35
being a total victory.
14:35
being a total victory. >> Okay, let's hear it.
14:36
>> Okay, let's hear it.
14:36
>> Okay, let's hear it. >> Why only 75%?
14:37
>> Why only 75%?
14:37
>> Why only 75%? >> Ah, yes.
14:38
>> Ah, yes.
14:38
>> Ah, yes. >> If the AI successfully found the glitch,
14:41
>> If the AI successfully found the glitch,
14:41
>> If the AI successfully found the glitch, right, and the university synthesized
14:43
right, and the university synthesized
14:43
right, and the university synthesized the vaccine to train the immune system
14:45
the vaccine to train the immune system
14:45
the vaccine to train the immune system to attack that specific glitch.
14:47
to attack that specific glitch.
14:47
to attack that specific glitch. >> Yeah.
14:48
>> Yeah.
14:48
>> Yeah. >> What happened to the other 25% of the
14:49
>> What happened to the other 25% of the
14:49
>> What happened to the other 25% of the tumor?
14:50
tumor?
14:50
tumor? >> It's a great question.
14:51
>> It's a great question.
14:51
>> It's a great question. >> Why didn't the T-cell army wipe the
14:52
>> Why didn't the T-cell army wipe the
14:52
>> Why didn't the T-cell army wipe the whole thing out?
14:53
whole thing out?
14:53
whole thing out? >> Because tumor heterogeneity is one of
14:55
>> Because tumor heterogeneity is one of
14:55
>> Because tumor heterogeneity is one of the most complex challenges in oncology
14:57
the most complex challenges in oncology
14:57
the most complex challenges in oncology today.
14:57
today.
14:58
today. >> Heterogeneity, meaning it's not all the
15:00
>> Heterogeneity, meaning it's not all the
15:00
>> Heterogeneity, meaning it's not all the same.
15:00
same.
15:01
same. >> Exactly. We often mistakenly assume a
15:03
>> Exactly. We often mistakenly assume a
15:03
>> Exactly. We often mistakenly assume a tumor is just a uniform mass of
15:05
tumor is just a uniform mass of
15:06
tumor is just a uniform mass of identical cloned cell.
15:07
identical cloned cell.
15:07
identical cloned cell. >> Like a solid block of the exact same bad
15:09
>> Like a solid block of the exact same bad
15:09
>> Like a solid block of the exact same bad code.
15:10
code.
15:10
code. >> Right, but it's not.
15:12
>> Right, but it's not.
15:12
>> Right, but it's not. As a tumor grows and divides, it
15:14
As a tumor grows and divides, it
15:14
As a tumor grows and divides, it continues to mutate internally,
15:16
continues to mutate internally,
15:16
continues to mutate internally, >> Oh.
15:16
>> Oh.
15:16
>> Oh. >> creating these distinct subclonal
15:18
>> creating these distinct subclonal
15:18
>> creating these distinct subclonal populations.
15:19
populations.
15:19
populations. >> So it's mutating while it's growing.
15:21
>> So it's mutating while it's growing.
15:21
>> So it's mutating while it's growing. >> Yes.
15:22
>> Yes.
15:22
>> Yes. >> That remaining 25% represents a
15:24
>> That remaining 25% represents a
15:24
>> That remaining 25% represents a subpopulation of cancer cells within
15:26
subpopulation of cancer cells within
15:26
subpopulation of cancer cells within Rose's tumor that possessed a slightly
15:28
Rose's tumor that possessed a slightly
15:28
Rose's tumor that possessed a slightly different genetic profile than the
15:30
different genetic profile than the
15:30
different genetic profile than the dominant mass.
15:31
dominant mass.
15:31
dominant mass. >> Ah, I see.
15:32
>> Ah, I see.
15:32
>> Ah, I see. >> The vaccine was perfectly engineered to
15:33
>> The vaccine was perfectly engineered to
15:33
>> The vaccine was perfectly engineered to wipe out the main strain of the cancer,
15:36
wipe out the main strain of the cancer,
15:36
wipe out the main strain of the cancer, but those minor sub-factions evaded
15:38
but those minor sub-factions evaded
15:38
but those minor sub-factions evaded detection.
15:38
detection.
15:38
detection. >> Because their specific genetic disguise
15:41
>> Because their specific genetic disguise
15:41
>> Because their specific genetic disguise wasn't on the wanted poster the RNA
15:42
wasn't on the wanted poster the RNA
15:42
wasn't on the wanted poster the RNA delivered.
15:43
delivered.
15:43
delivered. >> That analogy captures the mechanics
15:45
>> That analogy captures the mechanics
15:45
>> That analogy captures the mechanics perfectly.
15:45
perfectly.
15:46
perfectly. >> So, the tumor is almost like a massive
15:47
>> So, the tumor is almost like a massive
15:47
>> So, the tumor is almost like a massive criminal syndicate with different
15:49
criminal syndicate with different
15:49
criminal syndicate with different factions.
15:49
factions.
15:49
factions. >> Yes, exactly.
15:50
>> Yes, exactly.
15:50
>> Yes, exactly. >> You gave the immune system the profile
15:52
>> You gave the immune system the profile
15:52
>> You gave the immune system the profile for the main boss and their crew, and
15:54
for the main boss and their crew, and
15:54
for the main boss and their crew, and they took them out, but a splinter group
15:56
they took them out, but a splinter group
15:56
they took them out, but a splinter group with a completely different disguise
15:58
with a completely different disguise
15:58
with a completely different disguise survived the raid.
15:59
survived the raid.
15:59
survived the raid. >> That's exactly what happened, and this
16:01
>> That's exactly what happened, and this
16:01
>> That's exactly what happened, and this limitation shows exactly where the
16:03
limitation shows exactly where the
16:03
limitation shows exactly where the future of this technology is heading.
16:05
future of this technology is heading.
16:05
future of this technology is heading. >> Oh, where is it heading?
16:07
>> Oh, where is it heading?
16:07
>> Oh, where is it heading? >> The next evolution is multivalent or
16:10
>> The next evolution is multivalent or
16:10
>> The next evolution is multivalent or combination vaccines.
16:12
combination vaccines.
16:12
combination vaccines. >> Meaning more than one target.
16:13
>> Meaning more than one target.
16:13
>> Meaning more than one target. >> Yes, instead of the AI just looking for
16:16
>> Yes, instead of the AI just looking for
16:16
>> Yes, instead of the AI just looking for the one dominant mutation, the
16:18
the one dominant mutation, the
16:18
the one dominant mutation, the computational power would be used to
16:19
computational power would be used to
16:19
computational power would be used to identify three, four, or five different
16:22
identify three, four, or five different
16:22
identify three, four, or five different genetic signatures representing all the
16:24
genetic signatures representing all the
16:24
genetic signatures representing all the various sub-populations within the
16:26
various sub-populations within the
16:26
various sub-populations within the tumor.
16:26
tumor.
16:26
tumor. >> So, rounding up all the bosses.
16:28
>> So, rounding up all the bosses.
16:28
>> So, rounding up all the bosses. >> Exactly. The lab would have then print a
16:31
>> Exactly. The lab would have then print a
16:31
>> Exactly. The lab would have then print a single RNA vaccine containing multiple
16:33
single RNA vaccine containing multiple
16:33
single RNA vaccine containing multiple different blueprints.
16:34
different blueprints.
16:34
different blueprints. >> Wow.
16:35
>> Wow.
16:35
>> Wow. >> Simultaneously cutting off every
16:37
>> Simultaneously cutting off every
16:37
>> Simultaneously cutting off every evolutionary escape route the cancer
16:38
evolutionary escape route the cancer
16:39
evolutionary escape route the cancer might try to take.
16:40
might try to take.
16:40
might try to take. >> That is just brilliant. So, what does
16:42
>> That is just brilliant. So, what does
16:42
>> That is just brilliant. So, what does this all mean for us because as much as
16:44
this all mean for us because as much as
16:44
this all mean for us because as much as we love a heartwarming story about a guy
16:46
we love a heartwarming story about a guy
16:46
we love a heartwarming story about a guy using tech to save his best friend
16:48
using tech to save his best friend
16:48
using tech to save his best friend >> Which is a great story.
16:50
>> Which is a great story.
16:50
>> Which is a great story. >> It has massive implications. Rose's
16:52
>> It has massive implications. Rose's
16:52
>> It has massive implications. Rose's success is essentially a pilot study.
16:54
success is essentially a pilot study.
16:54
success is essentially a pilot study. It's a critical proof of concept for
16:56
It's a critical proof of concept for
16:56
It's a critical proof of concept for something called the One Health
16:57
something called the One Health
16:57
something called the One Health blueprint.
16:58
blueprint.
16:58
blueprint. >> Right, the One Health approach is a
17:00
>> Right, the One Health approach is a
17:00
>> Right, the One Health approach is a vital framework in modern medical
17:01
vital framework in modern medical
17:01
vital framework in modern medical research. It recognizes that human
17:03
research. It recognizes that human
17:04
research. It recognizes that human health, animal health, and environmental
17:06
health, animal health, and environmental
17:06
health, animal health, and environmental health are inextricably linked.
17:08
health are inextricably linked.
17:08
health are inextricably linked. >> We're all in the same ecosystem.
17:09
>> We're all in the same ecosystem.
17:09
>> We're all in the same ecosystem. >> Exactly. Dogs serve as the perfect
17:12
>> Exactly. Dogs serve as the perfect
17:12
>> Exactly. Dogs serve as the perfect biological sentinels for human cancers.
17:14
biological sentinels for human cancers.
17:14
biological sentinels for human cancers. >> Sentinels because they live with us.
17:17
>> Sentinels because they live with us.
17:17
>> Sentinels because they live with us. >> Yeah, they live in our exact
17:18
>> Yeah, they live in our exact
17:18
>> Yeah, they live in our exact environments, they drink our water, they
17:20
environments, they drink our water, they
17:20
environments, they drink our water, they are exposed to the same household
17:22
are exposed to the same household
17:22
are exposed to the same household chemicals, and most importantly, their
17:24
chemicals, and most importantly, their
17:24
chemicals, and most importantly, their immune systems develop remarkably
17:27
immune systems develop remarkably
17:27
immune systems develop remarkably similar spontaneous cancers to humans.
17:30
similar spontaneous cancers to humans.
17:30
similar spontaneous cancers to humans. >> The spontaneous cancers meaning not
17:33
>> The spontaneous cancers meaning not
17:33
>> The spontaneous cancers meaning not genetically engineered in a lab?
17:34
genetically engineered in a lab?
17:34
genetically engineered in a lab? >> Right, and they respond to treatments in
17:36
>> Right, and they respond to treatments in
17:36
>> Right, and they respond to treatments in very similar ways to humans, both
17:38
very similar ways to humans, both
17:38
very similar ways to humans, both genetically and biologically.
17:40
genetically and biologically.
17:40
genetically and biologically. >> Wait, I'm stuck on something here.
17:41
>> Wait, I'm stuck on something here.
17:41
>> Wait, I'm stuck on something here. >> Sure.
17:42
>> Sure.
17:42
>> Sure. >> I get how a dog's tumor is genetically
17:44
>> I get how a dog's tumor is genetically
17:44
>> I get how a dog's tumor is genetically similar to a human's, and the
17:46
similar to a human's, and the
17:46
similar to a human's, and the environmental factors overlap. But, how
17:48
environmental factors overlap. But, how
17:48
environmental factors overlap. But, how does curing a dog actually fast-track
17:51
does curing a dog actually fast-track
17:51
does curing a dog actually fast-track human trials in any legal or scientific
17:53
human trials in any legal or scientific
17:53
human trials in any legal or scientific way?
17:54
way?
17:54
way? >> It's about the regulatory pathway.
17:56
>> It's about the regulatory pathway.
17:56
>> It's about the regulatory pathway. >> Because aren't the FDA regulations and
17:58
>> Because aren't the FDA regulations and
17:58
>> Because aren't the FDA regulations and human clinical trial pathway of
17:59
human clinical trial pathway of
17:59
human clinical trial pathway of completely separate from veterinary
18:01
completely separate from veterinary
18:01
completely separate from veterinary medicine? A success in a vet clinic
18:03
medicine? A success in a vet clinic
18:03
medicine? A success in a vet clinic doesn't magically approve a drug for a
18:05
doesn't magically approve a drug for a
18:05
doesn't magically approve a drug for a human hospital.
18:06
human hospital.
18:06
human hospital. >> You are absolutely correct that the
18:08
>> You are absolutely correct that the
18:08
>> You are absolutely correct that the regulatory bodies are separate, but you
18:10
regulatory bodies are separate, but you
18:10
regulatory bodies are separate, but you have to look at the mechanism of
18:12
have to look at the mechanism of
18:12
have to look at the mechanism of comparative oncology.
18:13
comparative oncology.
18:13
comparative oncology. >> Okay, break that down for me.
18:15
>> Okay, break that down for me.
18:15
>> Okay, break that down for me. >> When researchers apply to the FDA for an
18:17
>> When researchers apply to the FDA for an
18:17
>> When researchers apply to the FDA for an investigational new drug application, an
18:19
investigational new drug application, an
18:19
investigational new drug application, an IND, to begin human trials, they need
18:22
IND, to begin human trials, they need
18:22
IND, to begin human trials, they need massive amounts of preclinical safety
18:25
massive amounts of preclinical safety
18:25
massive amounts of preclinical safety and efficacy data.
18:27
and efficacy data.
18:27
and efficacy data. >> They need proof it won't just kill the
18:28
>> They need proof it won't just kill the
18:28
>> They need proof it won't just kill the patient.
18:29
patient.
18:29
patient. >> Exactly. Now, traditionally, this data
18:32
>> Exactly. Now, traditionally, this data
18:32
>> Exactly. Now, traditionally, this data comes from engineered mice.
18:33
comes from engineered mice.
18:33
comes from engineered mice. >> Lab mice.
18:34
>> Lab mice.
18:34
>> Lab mice. >> Right, which are honestly terrible
18:36
>> Right, which are honestly terrible
18:36
>> Right, which are honestly terrible proxies for complex human immune
18:38
proxies for complex human immune
18:38
proxies for complex human immune responses.
18:39
responses.
18:39
responses. >> Because they're mice.
18:40
>> Because they're mice.
18:40
>> Because they're mice. >> Yes, but by proving that this specific
18:43
>> Yes, but by proving that this specific
18:43
>> Yes, but by proving that this specific RNA manufacturing pipeline and the LNP
18:45
RNA manufacturing pipeline and the LNP
18:45
RNA manufacturing pipeline and the LNP delivery system works safely and
18:47
delivery system works safely and
18:47
delivery system works safely and effectively in dogs.
18:48
effectively in dogs.
18:48
effectively in dogs. >> Dogs who have complex natural immune
18:50
>> Dogs who have complex natural immune
18:50
>> Dogs who have complex natural immune systems.
18:51
systems.
18:51
systems. >> Exactly, dogs battling real spontaneous
18:54
>> Exactly, dogs battling real spontaneous
18:54
>> Exactly, dogs battling real spontaneous tumors. When you prove it there,
18:56
tumors. When you prove it there,
18:56
tumors. When you prove it there, researchers generate incredibly
18:58
researchers generate incredibly
18:58
researchers generate incredibly high-weight preclinical data.
18:59
high-weight preclinical data.
18:59
high-weight preclinical data. >> Oh, I see.
19:00
>> Oh, I see.
19:00
>> Oh, I see. >> The FDA doesn't accept the dog cure as a
19:03
>> The FDA doesn't accept the dog cure as a
19:03
>> The FDA doesn't accept the dog cure as a human cure, but the overwhelming proof
19:05
human cure, but the overwhelming proof
19:05
human cure, but the overwhelming proof of the mechanism's safety in dogs
19:07
of the mechanism's safety in dogs
19:07
of the mechanism's safety in dogs dramatically accelerates the regulatory
19:09
dramatically accelerates the regulatory
19:09
dramatically accelerates the regulatory approval process to begin human trials.
19:11
approval process to begin human trials.
19:11
approval process to begin human trials. >> That makes total sense. The mechanism
19:13
>> That makes total sense. The mechanism
19:13
>> That makes total sense. The mechanism itself is what is being proven.
19:14
itself is what is being proven.
19:14
itself is what is being proven. >> Precisely.
19:15
>> Precisely.
19:15
>> Precisely. >> And it doesn't even stop at cancer, does
19:17
>> And it doesn't even stop at cancer, does
19:17
>> And it doesn't even stop at cancer, does it?
19:18
it?
19:18
it? Professor Thordarson speculated that
19:20
Professor Thordarson speculated that
19:20
Professor Thordarson speculated that this exact same RNA platform could be
19:23
this exact same RNA platform could be
19:23
this exact same RNA platform could be adapted for completely different
19:24
adapted for completely different
19:24
adapted for completely different conditions.
19:25
conditions.
19:25
conditions. >> It can.
19:26
>> It can.
19:26
>> It can. The manufacturing pipeline for RNA is
19:28
The manufacturing pipeline for RNA is
19:28
The manufacturing pipeline for RNA is fundamentally disease agnostic.
19:30
fundamentally disease agnostic.
19:30
fundamentally disease agnostic. >> Disease agnostic, meaning it doesn't
19:32
>> Disease agnostic, meaning it doesn't
19:32
>> Disease agnostic, meaning it doesn't care what it's treating.
19:33
care what it's treating.
19:33
care what it's treating. >> Right. The machinery printing the mRNA
19:36
>> Right. The machinery printing the mRNA
19:36
>> Right. The machinery printing the mRNA strands doesn't care what code it is
19:38
strands doesn't care what code it is
19:38
strands doesn't care what code it is assembling.
19:39
assembling.
19:39
assembling. Once you have the established laboratory
19:41
Once you have the established laboratory
19:41
Once you have the established laboratory infrastructure to safely print and
19:43
infrastructure to safely print and
19:43
infrastructure to safely print and package custom RNA, you simply swap out
19:46
package custom RNA, you simply swap out
19:46
package custom RNA, you simply swap out the digital payload.
19:47
the digital payload.
19:47
the digital payload. >> Like changing a software cartridge.
19:48
>> Like changing a software cartridge.
19:49
>> Like changing a software cartridge. >> Exactly.
19:50
>> Exactly.
19:50
>> Exactly. Today, we are talking about a cancer
19:52
Today, we are talking about a cancer
19:52
Today, we are talking about a cancer blueprint. Tomorrow, that exact same
19:55
blueprint. Tomorrow, that exact same
19:55
blueprint. Tomorrow, that exact same platform could be utilized to deliver
19:56
platform could be utilized to deliver
19:56
platform could be utilized to deliver genetic instructions to correct
19:58
genetic instructions to correct
19:58
genetic instructions to correct anomalies driving complex neurological
20:00
anomalies driving complex neurological
20:00
anomalies driving complex neurological diseases.
20:01
diseases.
20:01
diseases. >> Like Alzheimer's or Parkinson's.
20:03
>> Like Alzheimer's or Parkinson's.
20:03
>> Like Alzheimer's or Parkinson's. >> Potentially, yes. Or even to modulate
20:05
>> Potentially, yes. Or even to modulate
20:05
>> Potentially, yes. Or even to modulate the immune system for severe autoimmune
20:07
the immune system for severe autoimmune
20:07
the immune system for severe autoimmune disorders. The physical delivery system
20:10
disorders. The physical delivery system
20:10
disorders. The physical delivery system remains identical. Only the software
20:12
remains identical. Only the software
20:12
remains identical. Only the software being loaded into it changes.
20:13
being loaded into it changes.
20:13
being loaded into it changes. >> That is wild. It really highlights the
20:15
>> That is wild. It really highlights the
20:15
>> That is wild. It really highlights the rise of the citizen scientist in all of
20:17
rise of the citizen scientist in all of
20:17
rise of the citizen scientist in all of this.
20:17
this.
20:17
this. >> It absolutely does.
20:19
>> It absolutely does.
20:19
>> It absolutely does. >> Because we are seeing highly motivated
20:21
>> Because we are seeing highly motivated
20:21
>> Because we are seeing highly motivated individuals with tech backgrounds
20:23
individuals with tech backgrounds
20:23
individuals with tech backgrounds looking at these historically closed-off
20:25
looking at these historically closed-off
20:25
looking at these historically closed-off medical systems and basically forcing a
20:29
medical systems and basically forcing a
20:29
medical systems and basically forcing a faster, more transparent evolution of
20:30
faster, more transparent evolution of
20:30
faster, more transparent evolution of health care.
20:31
health care.
20:31
health care. >> They're pushing the boundaries.
20:32
>> They're pushing the boundaries.
20:32
>> They're pushing the boundaries. >> They are demanding that traditional
20:34
>> They are demanding that traditional
20:34
>> They are demanding that traditional medical frameworks adapt to the speed of
20:36
medical frameworks adapt to the speed of
20:36
medical frameworks adapt to the speed of the digital age.
20:37
the digital age.
20:37
the digital age. >> But the democratization of medical
20:39
>> But the democratization of medical
20:39
>> But the democratization of medical research, while it is accelerating
20:41
research, while it is accelerating
20:41
research, while it is accelerating discovery, it is crucial to temper that
20:44
discovery, it is crucial to temper that
20:44
discovery, it is crucial to temper that excitement with a clear understanding of
20:46
excitement with a clear understanding of
20:46
excitement with a clear understanding of the risks.
20:46
the risks.
20:47
the risks. >> We can't just have everyone playing
20:48
>> We can't just have everyone playing
20:48
>> We can't just have everyone playing doctor.
20:48
doctor.
20:48
doctor. >> Right. The collaboration between
20:50
>> Right. The collaboration between
20:50
>> Right. The collaboration between motivated families and local research
20:52
motivated families and local research
20:52
motivated families and local research labs can be incredibly productive. But
20:55
labs can be incredibly productive. But
20:55
labs can be incredibly productive. But it requires strict adherence to safety
20:57
it requires strict adherence to safety
20:57
it requires strict adherence to safety protocols.
20:58
protocols.
20:58
protocols. >> Which is exactly where we should land
21:00
>> Which is exactly where we should land
21:00
>> Which is exactly where we should land this. Because if you are listening to
21:02
this. Because if you are listening to
21:02
this. Because if you are listening to this right now, and you have a pet
21:03
this right now, and you have a pet
21:03
this right now, and you have a pet facing a terrifying health crisis,
21:06
facing a terrifying health crisis,
21:06
facing a terrifying health crisis, >> which is an awful place to
21:07
>> which is an awful place to
21:07
>> which is an awful place to >> It's so easy to want to jump online,
21:09
>> It's so easy to want to jump online,
21:09
>> It's so easy to want to jump online, fire up ChatGPT, and try to replicate
21:12
fire up ChatGPT, and try to replicate
21:12
fire up ChatGPT, and try to replicate this tomorrow.
21:13
this tomorrow.
21:13
this tomorrow. >> Too easy, really.
21:14
>> Too easy, really.
21:14
>> Too easy, really. >> Yeah. But, we need to look at the
21:15
>> Yeah. But, we need to look at the
21:15
>> Yeah. But, we need to look at the practical steps and safety protocols
21:18
practical steps and safety protocols
21:18
practical steps and safety protocols based on the realities of this specific
21:20
based on the realities of this specific
21:20
based on the realities of this specific source material.
21:21
source material.
21:21
source material. >> So, the first and most critical step is
21:23
>> So, the first and most critical step is
21:23
>> So, the first and most critical step is acknowledging that what happened with
21:25
acknowledging that what happened with
21:25
acknowledging that what happened with Rose is currently a one-off, highly
21:27
Rose is currently a one-off, highly
21:27
Rose is currently a one-off, highly experimental treatment.
21:29
experimental treatment.
21:29
experimental treatment. >> It's not the norm.
21:29
>> It's not the norm.
21:29
>> It's not the norm. >> No. Custom synthesized RNA vaccines are
21:32
>> No. Custom synthesized RNA vaccines are
21:33
>> No. Custom synthesized RNA vaccines are not something you can just order off a
21:34
not something you can just order off a
21:34
not something you can just order off a menu at your local veterinary clinic. It
21:36
menu at your local veterinary clinic. It
21:36
menu at your local veterinary clinic. It is entirely outside the realm of
21:38
is entirely outside the realm of
21:38
is entirely outside the realm of standard care.
21:39
standard care.
21:39
standard care. >> Right. Step two is you must start by
21:42
>> Right. Step two is you must start by
21:42
>> Right. Step two is you must start by consulting your licensed veterinarian.
21:44
consulting your licensed veterinarian.
21:44
consulting your licensed veterinarian. >> Always.
21:45
>> Always.
21:45
>> Always. >> You need to ask them specific questions
21:46
>> You need to ask them specific questions
21:47
>> You need to ask them specific questions about the limits, the potential
21:48
about the limits, the potential
21:48
about the limits, the potential benefits, and the absolute reality of
21:51
benefits, and the absolute reality of
21:51
benefits, and the absolute reality of traditional care for your specific pet's
21:54
traditional care for your specific pet's
21:54
traditional care for your specific pet's condition before you look anywhere else.
21:56
condition before you look anywhere else.
21:56
condition before you look anywhere else. >> You have to know what the standard
21:57
>> You have to know what the standard
21:57
>> You have to know what the standard baseline is before you deviate from it.
21:59
baseline is before you deviate from it.
21:59
baseline is before you deviate from it. >> Exactly. And step three, if you do
22:01
>> Exactly. And step three, if you do
22:01
>> Exactly. And step three, if you do decide to explore experimental research
22:03
decide to explore experimental research
22:04
decide to explore experimental research paths, you cannot do it alone.
22:06
paths, you cannot do it alone.
22:06
paths, you cannot do it alone. >> Never alone.
22:06
>> Never alone.
22:06
>> Never alone. >> As we discussed, the AI is a flawed
22:08
>> As we discussed, the AI is a flawed
22:08
>> As we discussed, the AI is a flawed tool.
22:09
tool.
22:09
tool. >> Very flawed. You must coordinate
22:10
>> Very flawed. You must coordinate
22:10
>> Very flawed. You must coordinate effectively with local research
22:12
effectively with local research
22:12
effectively with local research universities and academic institutions.
22:13
universities and academic institutions.
22:14
universities and academic institutions. >> You need that safety net.
22:15
>> You need that safety net.
22:15
>> You need that safety net. >> Yes. The clinical oversight, the
22:17
>> Yes. The clinical oversight, the
22:17
>> Yes. The clinical oversight, the biological error checking, and the
22:19
biological error checking, and the
22:19
biological error checking, and the sterile manufacturing provided by a
22:21
sterile manufacturing provided by a
22:21
sterile manufacturing provided by a university lab are non-negotiable
22:23
university lab are non-negotiable
22:23
university lab are non-negotiable requirements for safely turning a
22:25
requirements for safely turning a
22:25
requirements for safely turning a digital idea into a biological reality.
22:27
digital idea into a biological reality.
22:27
digital idea into a biological reality. >> It's all about balancing that
22:29
>> It's all about balancing that
22:29
>> It's all about balancing that tech-driven hope with rigorous
22:31
tech-driven hope with rigorous
22:31
tech-driven hope with rigorous scientific safety.
22:32
scientific safety.
22:32
scientific safety. >> Beautifully said. And if we look toward
22:34
>> Beautifully said. And if we look toward
22:34
>> Beautifully said. And if we look toward the horizon, there's a final thought I
22:36
the horizon, there's a final thought I
22:36
the horizon, there's a final thought I want to leave you with today.
22:37
want to leave you with today.
22:37
want to leave you with today. >> Let's hear it.
22:38
>> Let's hear it.
22:38
>> Let's hear it. >> If a tech guy using an off-the-shelf AI
22:41
>> If a tech guy using an off-the-shelf AI
22:41
>> If a tech guy using an off-the-shelf AI tool and just $3,000
22:43
tool and just $3,000
22:43
tool and just $3,000 can successfully hack a custom genetic
22:46
can successfully hack a custom genetic
22:46
can successfully hack a custom genetic cancer vaccine for his dog,
22:48
cancer vaccine for his dog,
22:48
cancer vaccine for his dog, what happens to the massive,
22:50
what happens to the massive,
22:50
what happens to the massive, multi-billion-dollar global
22:52
multi-billion-dollar global
22:52
multi-billion-dollar global pharmaceutical industry when the tools
22:54
pharmaceutical industry when the tools
22:54
pharmaceutical industry when the tools to create highly personalized individual
22:56
to create highly personalized individual
22:56
to create highly personalized individual cures are eventually sitting on
22:58
cures are eventually sitting on
22:58
cures are eventually sitting on everyone's desktop?
22:59
everyone's desktop?
22:59
everyone's desktop? >> For more information, visit the original
23:00
>> For more information, visit the original
23:00
>> For more information, visit the original source article by Dog Agility Magazine
23:02
source article by Dog Agility Magazine
23:02
source article by Dog Agility Magazine at dogagility.com.