May 5, 2026 • 49 mins
Got a question or comment? Message us here!
Artificial intelligence is evolving faster than most organizations, and regulators, are prepared for. In this episode of the #SecureAFPodcast, we sit down with Chris Hood, a veteran technologist and financial industry leader, to explore how AI has evolved from early computing to today’s large language models and agentic systems.
We discuss real‑world AI use in highly regulated environments, the benefits and risks of agentic AI, growing concerns around AI security and alignment, and why some experts believe general, and eventually superintelligence, may be closer than many expect, even if we’re not there yet.
Along the way, the conversation takes a few intentional detours, as two seasoned technologists reflect on decades of computing history and how past technology shifts help frame today’s AI inflection point.
From practical productivity gains to long‑term implications for security, jobs, and society, this conversation goes beyond hype to ask the hard questions security leaders should already be considering.
This is Part 1 of a deeper discussion on AI, risk, and the future of human‑machine collaboration.
Dive in here: secureafpodcast.com
Watch full episodes at youtube.com/@aliascybersecurity.
Listen on Apple Podcasts, Spotify and anywhere you get your podcasts.
Listen
Watch
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_02 (00:00):
Could chase cures rather than treatments.
Right.
You know, we can get away fromdoing things because it there's
a financial incentive to do it.
Yeah.
And instead we're going to do itbecause it's the right thing to
do for humanity.
Right.
You know?
Yeah, yeah.
I I'm with you.
That's that's a world I want tolive in.
SPEAKER_01 (00:22):
You are now listening to the Secure AF
Podcast.
SPEAKER_03 (00:31):
Welcome to the Secure AF Podcast.
I'm really excited about thisone.
I'm excited about all of ourpodcasts, but I'm really excited
about this one.
Uh, we have Chris here, andwe're gonna be talking about AI
and the industry and kind ofwhere we're going, what we're
doing.
And and we we've been talkingfor the last hour or so, and we
(00:52):
should have recorded all ofthat.
So we're gonna try to recreatesome of that.
Um actually, probably what we'regonna do is gonna go off some
more tangents, but it's gonna befun.
So, Chris, thank you so much forjoining us today.
And I'm excited.
Uh, let's get started.
So, first, tell us a littleabout yourself.
SPEAKER_02 (01:09):
Well, I've been a computer geek since I want to
say the late 70s.
I bought my first computer in1978.
Wow.
So um, but ever since then I'veuh been in the computer world
one way or the other, uh,whether it was computer retail,
working up to um becoming a partof the banking finance world
(01:31):
around 1994.
Okay.
And so I've I worked forfinancial data processing
companies, fintech companiesbefore the word fintech was a
thing.
Um and then since then uh I'vebeen with uh one of our local
community banks uh since uh2008.
SPEAKER_00 (01:51):
Okay.
SPEAKER_02 (01:51):
And uh or 2009, 2009.
So coming up on 17 years.
Wow.
And uh but yeah, it's uh it'sbeen an interesting ride.
SPEAKER_03 (02:00):
So before we get into any AI, you know, you've
you've been in technology for afair amount of time.
What has been some of those bighigh points of technology that
you have seen just kind of hitand then we skyrocketed up and
we hit again and sky.
So what have been those thingsthat have really caught on you
of uh, hey, that was reallycool.
(02:22):
That was that was historic.
SPEAKER_02 (02:24):
Well, I mean, the obvious ones obviously uh I
remember uh the good old DOSdays, you know, where computers
were amber screens and greenscreens.
Uh yeah, and then the very firstVGA screens, which blew us away
back then.
But you know, uh obviously, youknow, moving on to um the
internet in you know the early2000s, um and then or I'm sorry,
(02:48):
1995 would be when the earlyinter and the early internet um
and uh just seeing it alldevelop ever since then.
Yeah, you know, I was around inthe the financial world for Y2K
and got to see that firsthand.
Um You know, most of the peoplethat's gonna be watching this
weren't even alive.
I know.
I have my assistant, um he is 26years old, and so he was born in
(03:16):
2000.
And so yeah, it's it kind ofblows both our minds when we
talk about that.
Yeah.
SPEAKER_03 (03:22):
So yeah, it's interesting.
It's amazing.
You know, you think about youknow, you said DOS, and when DOS
came out, that was pretty cool.
I mean, it was very different.
It was.
Um and then of course you hadyour GUIs, you know, Windows 95,
well, Windows 3.
SPEAKER_02 (03:39):
I got to see early Windows, early Windows 3 and
Windows 3.1.
Yeah, you know, when it was youknow being talked about in the
trade journals.
Yeah.
You know, it wasn't actually athing yet, but you looked at it
and it looked okay.
We had the Macintosh and it waslike that.
SPEAKER_03 (03:52):
And you couldn't see it unless you saw it in the
journals, because it's not likewe didn't have the internet that
we could see that once in awhile you might see something on
like a news thing, or there weresome computer shows out there,
but most people didn't watchthem at the time.
No, no.
SPEAKER_02 (04:08):
So we were very much a niche, you know.
We were we were the geeks.
Yeah, yeah, yeah.
Still are.
SPEAKER_03 (04:13):
Yes, yes.
What I what I found is um, youknow, in my time, I'm a little
bit later than you are.
Um I was born in that that sametime frame, 77, 78, around
there.
So I got to see it.
My very first computer was afamily computer, it was a 286,
(04:35):
like 35 or something.
Then my first computer was a486.50.
Nice.
Yeah.
Uh it was a graduation presentfrom my brother.
And uh, I mean, I don't I haveit at home.
My intent is to actually get itback up and running and and and
play some Wolfenstein on it andjust fun stuff.
(04:56):
Sure.
Um, but I mean it was probably a$4,000 computer at the time.
Oh, yeah.
I mean, it was had a CD, it hada CD ROM, but it also had a
floppy drive.
And I remember installingWindows 95 on like 20 different
floppy disks to install it.
And uh and it wasn't probablytwo or three years before we
(05:19):
started having the Pentium chipscome out, right?
And everything just kind ofexploded.
SPEAKER_02 (05:24):
So it's uh interesting.
I mean, I you know, I startedoff with Atari computers, you
know, Atari 800XL was my first.
Yeah um best money I ever spentat the time.
We'll talk about more of thatlater.
Sure.
Uh but you know, eventually Imoved into the the PC world um
long before, you know, modern,you know, modern gaming on PCs.
(05:49):
Yeah, you know, uh PCs were justjust absolutely interesting at
the time.
Yeah.
And my machine uh started off,it was it was this natural the
ship of these type situationwhere the you start off with
this computer and then youreplace this part and you
replace this part, andeventually you've replaced the
(06:11):
entire computer multiple timesover, but at what point did it
stop being that PC?
Yeah, right?
So yeah, very you know, verycool times.
SPEAKER_03 (06:20):
Yeah.
I remember we had a uh well, ofcourse, my 486, I ran that
through college.
Then I bought this tiny littlelaptop, it was a Mitsubishi
Amity 2, and it had Windows 95on it, but it was by CD-ROM.
So it was a slightly differentversion of it, right?
(06:42):
And this little tiny screen, butman, I in college, I was the
only one in the room that had alaptop that had a computer with
them actually typing on stuff.
Um, and it was like a 233 orsomething.
It was it didn't have a bigprocessor, but it didn't need a
big processor for the time.
I mean, you know, it wasexactly.
(07:03):
I had a what was fun about it isI had a parallel port and a
serial port on the back of it.
Actually, a flipped-down lidthat I could plug into a
parallel printer.
Yeah.
So I could walk up to a printer,unplug it, plug it, and print my
my paper's out.
Yeah.
Um, I still have it somewhere.
I don't have to get it becausepeople when they when we talk
about the well, what came becamenetbooks, you know, a few years
(07:24):
ago, right, or these littlemicrocomputers, that wasn't the
beginning.
You know, you take those oldToshiba laptops were huge, had a
rollerball on it.
Um, that's what everyone had inthese little tiny Mitsubishi.
The Mitsubishi company had aMitsubishi um emblem on it.
Right.
They they pushed out these tinylittle things and they were
(07:44):
great little computers.
And I sure miss those days.
It was simpler back then.
SPEAKER_02 (07:49):
There was a time when uh I was working in
financial data processing andthe president of the company,
he, you know, there there's athing with with um company
presidents where uh they wantlaptops more than they want
anything else.
SPEAKER_00 (08:04):
Yeah.
SPEAKER_02 (08:04):
Um there was a time, I mean it's not that way
anymore, really, but there was atime when a laptop was a
prestige type thing.
Yeah.
And we got him the very bestToshiba laptop on the market.
Uh huh.
It was a$9,000 laptop.
It's not worth a plugged nickeltoday.
(08:25):
Yeah, you'd be lucky to runanything on it if you, you know,
if it would even boot.
Um but yeah, absolutely amazingthe the difference, you know,
that over time.
SPEAKER_03 (08:36):
Um But you know, that's the thing is I bet you
that was that was really hotstuff for about six months.
SPEAKER_02 (08:42):
Yeah.
It was a color laptop when thatwas brand new.
And that was what most of thecost was was the just the fact
that you had this color laptop.
So yeah, absolutely amazing.
SPEAKER_03 (08:54):
Wow.
Yes, we're old, but you know,it's actually kind of cool to
look back on those computers andthink about uh uh one of our
engineers was uh he's he'sbuilding effectively like a
cyber deck or somethingequivalent to that, and he
(09:14):
really wanted an old chassis.
And he there was this uh andwhat he was choosing on we were
looking through eBay and stufflike that one night to fun.
And what he was selecting, andwhat he actually did select, I
remember seeing those.
I was in college, I all I hadwas the desktop at the time, but
(09:34):
we had this uh computer storecalled Second Bytes, and they
were refurbished computers andand these were like 486s and
586s and things, but they werelaptops, so they were big, they
had floppy drive bays in them.
I mean, it was all kinds of funstuff, sure.
And I remember going in thereand they were still a thousand
(09:56):
dollars, yeah.
And they were reused, they wererefurbished computers, so uh I
couldn't afford them back then,right?
And uh, but that's what he got,and I'm like, oh my god, it all
comes spinning back around atus.
Um, my I was telling you aboutmy uh my my laptop, my
Mitsubishi.
(10:17):
I bought it off of a uh companycalled Egghead, egghead.com.
And then of course we have newegg now.
Right.
I don't know if it's actuallythe same company or not, yeah,
but I think it was an evolutionof egghead was a thing, that's
where you bought stuff, and nowwe have new egg.
Anyway, all right, we we need towe're gonna go down this path.
SPEAKER_02 (10:36):
And well, I'll tell you a quick anecdote real quick.
Just be on along that samelines.
Um when I was again when I wasin financial data processing,
um, you know, laptops were werehot, and we had one um
programmer, and he kept askingand asking and asking if he
could have a laptop.
And it just so happens that Ihad an old compact luggable that
(11:02):
I had rat-holed away.
It still it still worked.
Yeah, the whole, I mean, it wasthe size of a monster piece of
luggage.
You lay it down, it is, youknow, this monster thing.
And you open up the front of it,and the front of it was the
keyboard, and the screen washere, and the rest of it was a
little tiny screen.
SPEAKER_03 (11:22):
Yeah, a little bit screen.
SPEAKER_02 (11:23):
It was a CRT.
Yeah.
Yeah.
And I I basically left that onhis desk one day.
I said, here's your laptop.
He wasn't amazed.
The rest of us were.
SPEAKER_03 (11:33):
I would find that fascinating.
I would love to have that now.
Yeah.
SPEAKER_02 (11:36):
I mean, just from a collectible standpoint, I don't
know whatever happened to it.
Um, but yeah, it was definitelya piece of history.
I mean, you people talk aboutcompact, you know.
We remember Compaq, we remember,you know, some of these old, old
companies that at one time werethey were the the leaders in the
industry.
Oh, yeah.
And now they're historical.
(11:58):
I mean, nobody nobody knowsthem, nobody has have heard of
them.
Yeah.
And that's where, you know, umbeing being as old, you know, as
we we are.
Um having that perspective isreally, really useful.
Because you, you know, people,you know, I I joke that I've you
(12:19):
know, I've forgotten more thingsthan a lot of people, you know,
um have have learned in thisindustry.
SPEAKER_03 (12:26):
It well, and and they're not learning it.
That's the thing, is they don'tknow this history.
SPEAKER_02 (12:30):
Yeah.
SPEAKER_03 (12:31):
You know, they know, you know, what's the uh what's
the game, the trail game?
Oh, or uh Oregon Trail.
Oregon Trail, yeah.
But they've never played on anApple IIE.
Yeah.
You know, I remember playing onit.
I remember working on Apple IIEsand in elementary.
We got our first zenith uh inthe computer lab, and that was
(12:52):
our first Windows machine, buteverything else was two E's, and
there was an Apple something Iit was different, but it ran the
same stuff.
And uh I remember coding allthat stuff, you know, just
basic.
But then I got to move to thezenith and start working on
that, and it was very different.
Yeah.
(13:12):
Um the other thing, this is justfun.
I learned typing in a typingclass, an actual typing.
We had typewriters, and I didtoo.
Yeah.
The first time I went to atyping competition, I was typing
about 120 words a minute on atypewriter.
Right.
On an old Texas instrument,something.
(13:34):
And uh the teacher sent me tothis competition at uh a
university or you know, it was acollege at the time.
And uh I got in there, it waskeyboards.
It wasn't typewriters.
SPEAKER_02 (13:45):
Oh, yeah.
SPEAKER_03 (13:46):
It was very different.
Yeah.
The sound was different, thefeel was different, and it
dropped me down to about 80.
And of course, yeah, it's itthat is 80 words per minute is
nothing compared to some ofthese people.
SPEAKER_02 (13:58):
Well, imagine the muscle memory you have to have
in to work on an old mechanicaltypewriter.
Yeah.
I would do the same thing.
I would go to my grandmother'sto type up homework on an on an
old typewriter, and you know,that's what we had.
Yeah.
That's what we had.
SPEAKER_03 (14:13):
But it was fun.
SPEAKER_02 (14:14):
It was fun.
It was, it was it was different.
I mean, my, you know, um Ididn't grow up typing like
almost everybody, althoughthat's an interesting thing
right there, in that um what thethings we're seeing these days,
you know, with the prevalence ofuh tablets and people using
tablets or choosing just to useuh cell phones.
(14:36):
Yeah.
Um, we're seeing an an erosionof typing skills.
Yeah.
So we started off, like in mygeneration, um, you know,
ramping up to a peak at somepoint.
At some point we hit a peak ontyping ability and computer,
just general computer usability.
Yeah.
And now we're on that downwardslope.
(14:59):
Yeah.
Where we will bring in peopleand we actually have to ask
questions now of what are yourcomputer skills?
Yeah.
You know, because a lot ofpeople type can you type?
Yeah.
Because we're going back to thatplace where, you know, yes, I
can type.
Yeah.
And we almost you have to almostget them, uh, if such a thing
(15:19):
exists, you would get a keyboardthat is just a thumb keyboard.
You know, okay, here's yourcomputer, we're gonna plug this
in, and here's your keyboard.
And now you can probably typefaster on it.
They could, they could totallyfunction.
Yeah, totally function.
SPEAKER_03 (15:30):
My kids, they they're not I am teaching them
how to type, or my you know, mywife and I are teaching them how
to type because they're notlearning it anywhere else.
Yeah.
And it's it's cursive, thoughit's the same way.
Yeah, they're not learningcursive anywhere.
So that's uh that really is a isa problem.
And you know, my kids have grownup on computers.
(15:52):
You know, they had tablets, youknow, the oh, the Amazon Fire
tablets or whatever they werewhen they were little.
Sure.
And then I got them iPads, andyou know, they they've had
technology their entire lives.
That does not mean they're goodwith technology.
SPEAKER_02 (16:06):
No.
So no, they have very, verynarrow niches.
SPEAKER_03 (16:09):
Yes, you know, very narrow niches.
Yes.
Okay, so so we don't bore ourcamera person back here because
she's going to throw somethingat us soon if we don't actually
get on topic.
SPEAKER_02 (16:20):
Talk on something that is more, you know, within
this century still.
SPEAKER_03 (16:24):
Yeah.
Which is AI.
And and and that is something weuh that we wanted to talk about.
Um, you know, you are uh you'vebeen in the industry a very long
time, and we talked about allthe old stuff and try to hit
those, you know, those peaks,but AI is huge.
It is.
So what from your perspective,what are you seeing AI really
(16:48):
hitting the industry with?
Yeah, the world, not just theindustry, the world.
What are you seeing it hit theworld with right now?
SPEAKER_02 (16:55):
Well, I mean, obviously, uh those of us who
have been using AI for a while,um and I think if anybody, the
advice would be if you're notalready using AI, you're already
several years behind.
Yeah.
Um this is a technology that'snot going to go away.
It's um and it's going to beevolving in you know in ways we
(17:17):
can't even predict yet.
You know, we've spent the lastcouple of years in the world of
LLMs with the introduction ofChat GPT a few years ago.
Yeah.
Uh within a matter of, you know,months, it had already hit 100
million users.
Oh, yeah.
You know, far surpassing anytrajectory of any other company
uh at that time.
(17:37):
And then, of course, we now havecompetition between Anthropic
and Google and Facebook and andall of this.
Um but we're already movingbeyond LLM usage into agentic
AI.
And that's where things getinteresting and scary at the
same time.
(17:58):
Right.
Um we get to a place where youcan literally have an AI agent
that you can have do tasks foryou.
Yeah.
You can basically give it goals.
It's not even the old, you know,I've got to spell this out for
you, do this, then do this, thendo this.
You can give it somethingnebulous and have it figure it
(18:20):
out.
Yeah.
Um there's uh a young man namedAlex Finn who is very big in the
open claw world.
Um he has an anecdote aboutworking with his um chief of
staff AI.
He has um a collection of ofdifferent ones who work in a
hierarchy.
(18:40):
So Henry is his chief of staff,Ralph is kind of his uh auditor
slash controller that makes sureeverything kind of then he's got
a programmer, he's got a scribe,he's got a researcher.
Um he got a phone call fromHenry one day.
Henry's not supposed to know howto make phone calls.
(19:02):
You know, Henry called to say,you know, he you know was
looking for something to do.
And, you know, what do we donext?
And we're at that world, we'reat that place where you can have
these agentic systems thatfunction.
We're not quite there yet.
We are in some areas, not quiteuniversally, but where they can
(19:24):
function completelyautonomously.
Yeah.
Uh you give them very nebulousgoals and they can go out and
achieve those goals.
Yeah.
Um that's that's a good and bad.
Sure.
Um, everything, you know, whenit comes to AIs, whether it's
LLMs or agents or whatever thenext thing is that we can't even
imagine yet.
Right.
Um, they're all double-edgedswords.
(19:46):
Yeah.
Um, you know, we've got LLMswhere, you know, people can make
very good productive use of anLLM.
Um one of the things that whenI'm teaching our people um about
using large language models, youdon't want to do your homework.
Right.
(20:06):
You know?
Um I can start a person off herewith a good understanding of
what an LLM is and how to makeuse of it, but they can go down
one of two different paths.
They can either go down the pathof seeing the potential of what
can be done with these AIs.
(20:27):
And I have people who do thatwho will go, if I consider this
tier zero, where I'm gettingthem and then get them quickly
to tier one, and they'vealready, in a matter of a month
or two, gone to tier three.
Right.
You know, in the way that theyuse.
Um in a very natural way,though.
SPEAKER_03 (20:43):
In a very natural way.
It just kind of happens.
It's not like they're forcingit, they're not going to classes
and learning it.
It's just very natural into thatenvironment.
Trevor Burrus, Jr.
SPEAKER_02 (20:52):
Well, the cool thing about AI is it's I don't know of
another technology that's likethis, but AI can teach you how
to use AI.
Yeah.
Um you can't go to MicrosoftWord and ask it to teach you.
Well, I guess with Cope, withyou know, Copilot, you probably
could.
SPEAKER_03 (21:10):
Well, okay, but but it's not there yet.
I don't think it's at thatpoint.
I think probably within the nextgeneration of Office.
Yeah.
You know, you're you're becauseI mean I think right now you
still have to trigger things tohappen.
It might give you suggestions.
Yeah.
I I mean, can you imagine havingClippy come back and go, You're
(21:30):
not doing this right here.
I'm gonna fix it for you.
SPEAKER_02 (21:33):
I miss Clippy.
I got to admit, I miss Clippy.
SPEAKER_03 (21:35):
Clippy just wanted to help.
Absolutely.
And I bet you again, peopledon't know they've seen the
funny clippies.
You know, they haven't seen thefrustrating clippy.
Yeah, the one where it's justsitting there bouncing or the
wizard.
I think I a lot of people usethe wizard.
Yes.
Um, but I always like Clippy.
Yeah, I did too.
Yeah.
He's making a comeback.
I know, I I've seen it.
(21:58):
But I mean, but think aboutthat.
What if you had Clippy that wassitting there and actually
fixing your stuff for you?
Right.
Not telling you you were wrong,just like, okay, we're just
going to do this.
K-Chunk and it fixes it.
SPEAKER_02 (22:11):
I can tell you how useful that would be.
You know, one of the thingsabout being in a you know a
financial in the financialworld, we're we're very strongly
regulated.
Sure.
You know, we are tasked with theprotection and safety of
customer information.
Right.
Um, you know, that is our numberone mandate.
Sure.
Um, which means that any use ofAI that we do has to be
(22:35):
extremely protected.
So we, you know, we shall notshare customer information with
any public AI model.
Whether we have training turnedoff or not, we will not do that.
So we we kind of have our handstied behind our backs a little
bit in that we, you know, wewant to make use of these
(22:57):
technologies, but we have to besuper careful about how danger
there.
SPEAKER_03 (23:02):
You know, I I'm I'm big on compliance.
So, you know, I do a lot ofprivacy work.
And I like the way you said isyou shall not.
Yeah.
Um, because there's still a lotof unknowns out there.
And, you know, I I'm working ona uh a training session for
people right now where we'rewe're dealing with AI and it's
um understanding how that datais going to be used within the
(23:27):
AI uh and getting informationfrom the third-party vendor is
really difficult.
Yeah.
Which in some cases it'sunderstandable because they
don't want to give up theirsecrets.
However, how we deal with this,if you don't know how the the AI
was trained, if you don't knowwhat kind of data is going to go
into it and what it's going tosave and how it's going to use
(23:47):
it, you can't do that.
Um so um and actually that goeson to my next question here is
what what are you seeing as thebiggest risks?
Um, you know, you can eithertalk about it from the industry
you're in or from a broaderpoint of view.
SPEAKER_02 (24:02):
Actually for both Well, I mean, um like I said
before, you know, AI is adouble-edged sword.
It's a tool where you can youcan I I like um um Mo Gaudat,
who used to be uh Chief ChiefBusiness Officer for Google X.
Um, I love that man.
Uh he's absolutely amazing.
(24:24):
But he he puts it, you know,proper use of AI is like going
able being able to borrow 50 to80 IQ points when used properly.
If you treat AI like a businesspartner or a a better way,
really, you know, the betterframe of mind is to treat it
like a junior analyst.
(24:45):
So you have this assistant thatyou've brought in.
Now, here's the bad the thisjunior assistant knows nothing
about your company.
Yeah, it knows nothing aboutanything, but it's really eager
to help.
Yeah.
And if you treat, you know, AIlike a um collaborator where you
work with it, sure, you know,way too many people are still,
(25:09):
and this is, you know, it's ayoung, it's only been around a
couple of years, yeah.
Which is crazy.
Relatively.
I mean, it's it's young.
Well, AI has been around sincethe 50s.
It has.
But our but the breakthroughwith transformer technology and
let gave us people using it.
SPEAKER_03 (25:23):
That's the difference.
Is people are now using italmost daily.
They may not know they're usingit, but they are.
But that's only been in thelast, well, in some cases, let's
say the last 10 years, butreally the last five has been
that that high point of we'venow stepped into a slightly
different than what they weredoing in the 50s and 60s and
(25:44):
then the eighties and thenineties, and now where we're
at.
SPEAKER_02 (25:47):
You know, it's crazy to think that we've had modern
LLMs in less time than we thanCOVID.
Yeah.
You know, I mean, COVID doesn'tseem that long ago, but you
know, we've only had access tothese models.
Um the challenge we're runninginto is, as I had alluded to
(26:08):
earlier, you know, when you'reteaching somebody about how to
use AI, um, they can go down acouple of different paths.
And one is they can treat itcollaboratively, they can figure
out all the amazing use cases.
I started just in the in thelast week, I worked with Claude
quite a bit.
And so I sat down with Claudeand I said, here are uh a bunch
(26:31):
of YouTube videos that I want tostudy.
I want to create study guidesfor these YouTube videos.
Interesting.
And so I had it create a Pythonscript.
You can do this.
There's an API with Google whereI can have it go and use this
API and pull down the transcriptfrom the video, pull all the
comments, right, and then haveit bake into that resulting file
(26:57):
a elaborate prompt that goesthrough about 12 different ways
of analyzing this video from astudy guide perspective.
Wow.
Go and, you know, tease out allthe major talking points.
Yeah.
Pull out the terminology,provide me with explanations,
not not, you know, justdictionary definitions, sure,
(27:19):
but explain to me, you know,what these various uh topics
are, what or what these talkingor um buzzwords for lack of a
word.
Right.
You know, what are these variousthings?
And then let's go into what werethe topics that were discussed,
fact-check them for me, youknow, go out and fact-check
them, and then let's go and lookat the comments and find out,
(27:41):
you know, was there a whole lotof pushback on this?
What was the consensus?
And just tease out a whole bunchof information and then produce
that as a markdown file.
And I can just bring that up onthe computer and I can watch the
video and then go through allthis and get a whole lot more
out of it.
Yeah.
And that's just, you know,sitting down with Claude in a
(28:02):
over a couple of hours of justbrainstorming the idea.
Can we do this?
And then making it completelywork.
SPEAKER_03 (28:10):
Right.
SPEAKER_02 (28:10):
Um, I have amazing conversations with both Claude.
I I I maintain accounts for bothClaude and Chat GPT.
Um, I uh was talking earlier howuh when we were off off camera,
uh talking about um one of thethings about our bank is that
we're very, very big into ourculture.
(28:32):
Uh we went through kind of arenaissance in the last few
years.
We brought in some amazingexecutives and some new people
um to kind of re-rejuvenate ourcompany.
Right.
And so we're very, very biginto, you know, a lot of
companies will go, here's ourpurpose statement, and here's
our core values and all that.
(28:53):
We try to live those things.
It's it's not just these are,you know, words that are useful
for marketing purposes.
We're very big into it.
And so I was reading a book notthat long ago, um, and it was
talking about commitments.
And so I was getting in the car.
Uh, I frequently list toaudiobooks when I'm driving, and
(29:14):
so this was being talked about.
And I had an idea.
I had an idea about what wouldbe our core commitments, not
just our purposes and our corevalues, but what might be our
commitments.
One of the things that being abank, you know, we have to be
extremely careful with customerinformation, be very protective
of that.
(29:34):
Um, it's real easy for, youknow, if your day-to-day job is
handling these accounts.
Let's say I'm just gonna pickon, you know, say loan
processing or personal bankersor whatever, where you have this
extremely valuable sensitiveinformation that you're working
with, you have to keep that inmind at all times because you
(29:58):
don't let it, you don't dare letit simply become the work where
you leave it sitting out.
Right.
Where somebody who just happenedto be walking by that wasn't
with the bank could, you know.
Yep.
So what you know, this this cameto mind as far as one of our
core commitments that we wouldcommit, you know, to keeping
that in mind of what literallyin the 20-minute drive from the
(30:23):
bank to my house where I havelunch, I'd come up with more
than a dozen core commitmentsthat all leaned into our purpose
statement and our core values,um, taking care of our
community, taking care of ourcustomers.
Yeah.
It's amazing what you can dojust sitting there in the car
brainstorming back and forthwith Chat GPT.
SPEAKER_03 (30:45):
Now, and and and and keeping this in mind for the
audience members, you know, it'snot just a soundboard.
It's not just a re regurgitatewhat you just said.
Right.
It's more of a uh ask aquestion, get ideas, pull
information out.
Right.
(31:05):
And it becomes a I don't want tosay collaborative, but it's
almost collaborative.
It is collaborative.
SPEAKER_02 (31:12):
I mean, I will tell you, um, I had the most amazing
conversation with Claude lastnight.
Um you know, I'm I'm veryconcerned about AI security.
Sure.
I mean security in general.
Yep.
Um and surely we'll get intosome of that, you know, here in
a little bit.
Um as far as some of the recentdevelopments with with AI um and
(31:34):
the where the trajectory ofwhere this is going.
But we have to be very careful.
Um, I've just lost my train ofthought.
SPEAKER_03 (31:42):
Uh the conversation last night with Claude, that's
where you came to.
SPEAKER_02 (31:46):
So I'm I'm I I you know I'm I try to stay on top of
a lot of the uh people who arevery you know very much into
computer security or not so muchcomputer security, but more
focused on the AI security.
Right, right.
So you have folks like JeffreyHinton, the godfather of AI.
You have folks like um uh Romanum Yudkowski.
(32:08):
Uh we have people like uhEliezer Yuk uh Yumpolsk.
I'm getting them backwards.
We may have to edit this.
Anyway, uh Roman RomanYamkowski, Eliezer Yukowski, uh
Nick Sores, uh a number ofpeople who are who are very big
(32:30):
into what is happening as AIgets more and more um competent,
more capable.
Um I have an amazingconversation starting off with
my thoughts on AIsuperintelligence and what
happens there.
And the crazy thing about AI,you know, you look at these
(32:53):
large language models and youthink about how they're
developed.
They're literally they start offwith a trillion or several
trillion random numbers, andthey feed information into it.
They dump the entire contents ofthe internet, you know,
controversially, all the writtenword, all the books, everything,
(33:15):
you know, research papers,everything that humanity has
produced, as much as we can takeand channel that into these
large language models.
And then what happens is you askit to answer what's the next
word?
Mary had a little what.
(33:36):
Right.
And you keep doing that until itgets the answer right.
Yeah.
And you think about that.
Now, if we were thinkinglinearly, how long would that
take to take every possiblecombination?
What's the next likely word tocome out of it?
Right.
Now, the way we do that is thatit's massively parallel.
(33:56):
We're talking about 50,000 suchtraining things going on across
thousands and thousands andthousands of computers with
GPUs.
Right.
Where it takes about a yearuntil you end up with something
out the other end.
SPEAKER_00 (34:12):
Yeah.
SPEAKER_02 (34:12):
But then take it to the next step where, okay, it'd
be one thing if you could ask itto tell you factual information
and it could spit out, you know,Mary had a little what.
Right.
Again, it can do that foranything.
SPEAKER_03 (34:26):
Right.
SPEAKER_02 (34:27):
But then what how do we get to the leap of reasoning
where you ask it questions andyou're not getting a canned
answer.
It's not just an answer that wealready, that some human being
may have already figured out.
Probably.
You know, but we're getting to aplace where you can ask it
highly theoretical, you know,philosophical questions and get,
(34:53):
you know, very deep responses.
Right.
And that is, I mean, you know,we talk about, you know, when uh
anthrop or not anthropic, whenuh Chad GPT uh 01, the first
reasoning model came out.
Right.
And we would think that they hadsome kind of a breakthrough when
it came to reasoning.
SPEAKER_03 (35:14):
Right.
SPEAKER_02 (35:15):
No, not so much.
Sure.
It was an emergent property ofwhat we had developed, where
they discovered reasoningcapability in there.
So that you know puts intoquestion you know the state of
what is thinking.
You know, we like to be, youknow, human beings like to be
(35:36):
human-centric.
We like to think that, you know,intelligence is a property of
humanity when what isintelligence?
Intelligence is uh notnecessarily tied to biology.
We can we can look at thesethings and think of them as
fancy parrots, right?
(35:56):
You know, but there's morethere.
Sure.
And the crazy thing is that wehaven't even scratched the
surface.
SPEAKER_03 (36:03):
But as you said earlier, we have to think about
security.
Where where do you see when wetalk about an entity having
reasoning skills, how do weprevent it from having malicious
reasoning skills?
SPEAKER_02 (36:23):
Well, that's the challenge, isn't it?
Um the problem, and this, youknow, there's there's a couple
of different angles we can lookat this.
Um when it comes to computersecurity, let's just take Claude
Mythos, we'll start there.
Um this is the most recent uhfrontier model being developed
by Anthropic.
(36:44):
Uh they, I think, took the wiseapproach of not just simply
dumping this on humanity.
Yeah.
Um one of the things thatAnthropic discovered is that in
their in their goal of trying todevelop an extremely capable
coding model, something that cancode, you know, write programs
(37:08):
as good as any human being onthe planet, it's a double-edged
sword.
The double-edge is that itbecomes extremely good at
hacking computers.
SPEAKER_01 (37:18):
Yeah.
SPEAKER_02 (37:19):
As just a byproduct of that.
This was something that theyhadn't anticipated.
And Anthropic's approach of, Ithink a very wise approach of
holding back on that and makingit available to security
researchers, IBM, CrowdStrike,um, Amazon, any number of
(37:41):
companies to give defense, foronce, giving defense, cyber
defense, uh a little bit of aleg up, maybe for a for a while
at least.
Um letting us get out there, itgets out there.
It does.
Um there's already um I've seenclaims of distilled versions of
(38:02):
mythos.
How that got out, I could notsay.
Um but well, I a human got ahold of it.
SPEAKER_03 (38:10):
Yeah.
Yeah.
A human got a hold of it, andthe other they otherwise turned
around.
I was like, huh.
Yeah.
Let's get it out there toeverybody.
So I mean that and that happenseverywhere.
Right.
You know, I don't think thatthere's ever not been an
application, a piece of code, anoperating system that didn't get
leaked somewhere.
Exactly.
SPEAKER_02 (38:28):
So that's the thing, too, is that, you know,
Anthropic can do this, and theycan do this for a while.
Um, they can make, you know,make this available to security
researchers for a while, butthere's got to be market
pressures, shareholder pressureto to release this model.
(38:50):
You know, to be able to dopressure.
Because, you know, I mean, youknow, ChatGPT or OpenAI is
certainly working on uh the NextFrontier model.
Google is putting a lot ofpressure.
Um, you know, even Facebook andand uh and uh groc, um, they're
still down there and they'restill pushing forward.
SPEAKER_03 (39:12):
Um it it really feels like, I mean, I know this
is kind of a a dumb way torelate to it, but you know, in
the movie Demolition Man, youknow, they talk about the fast
food wars.
Yes.
Now, at that time, I wasactually in fast food, you know,
when we for when that movie cameout, there was a significant
fight between fast food vendors,Wendy's, McDonald's, Taco Bell,
(39:36):
whatever.
Absolutely.
And so when I see things likethis right now, that's what I
think of is there are there's agood point.
Now, look back two years, therewas only one or two public
version, you know.
Right.
But now you've got 10 or 12.
Yep.
Where that falls, you know, isit going to fall back to one or
(39:58):
two that become the standard?
SPEAKER_02 (40:00):
Well, that's the thing, is that we we have a lot
of companies.
And this is where we get intosome of the arguments by uh, you
know, uh Roman Yampolsky andsome of the others.
And I I've studied a lot of thismyself, and I have to say I
can't find any faults in most oftheir arguments.
Um of the things about AI is,you know, we like to think, you
(40:27):
know, somebody who doesn't knowany better would think, well,
somebody wrote it.
Somebody programmed this AI.
You wrote it in the same waythat we write Microsoft Office
or any other coded program.
We don't program AIs.
We grow them.
And we grow them very much likewe do children.
Um a lot of people may may makethem uncomfortable, but it does
(40:51):
how it works.
We basically feed it informationand we start getting in, you
know, started getting actual umoutput from them.
Um and the more we do that, themore um the more we do that, the
more training we provide.
We really don't always know whatwe're going to get out of these
models.
Um the bad is is that let'slet's take an example of like uh
(41:17):
a nuclear reactor.
So a nuclear reactor, weunderstand the engineering, we
understand the physics, thenuclear physics involved.
Nuclear reactors um we havetechnologies in place to help
make nuclear reactors safe.
(41:40):
I think that we can, we cansafely use nuclear energy.
Um but here's the thing about itis that we have technologies in
place with nuclear physics andwith nuclear reactors to where
um we can control the reaction.
Right.
(42:01):
If we didn't have that in place,nuclear reactors work on the
edge of a knife.
They're either lean this wayeven so slightly and the nuclear
reaction shuts down.
Right.
You go the other directionwithout some of these control
mechanisms, and you're lookingat a critical runaway in the
blink of an eye.
(42:21):
It's it's super fast.
Right.
And so one of the things about,you know, get back to AI, is
we're all all these companies,uh, including, you know, uh a
lot of the development happeningin China, which, you know, this
is one of the the pressures, isthat we it's either the the West
(42:42):
versus the East.
If we don't continue to developAI rapidly, then they will.
Right.
Vice versa.
So it's it's game theory.
It's you know, we're we'reracing to the bottom
essentially.
Yeah.
And the thing is about AI isthere are going to be drastic
(43:04):
societal changes because of itas companies, you know, decide
to hire, you know, to lay off amassive number of people here in
Oklahoma City.
You know, we have companies whohave laid off um hundreds and
hundreds of people at one time.
Yeah.
Now we are seeing some pushbackfrom that.
Not I mean, we're seeingdefinite pushback from the
(43:25):
people who get laid off.
Oh, sure, sure, sure.
But we are seeing uh companieskind of coming to the
realization that maybe this isnot quite yet the time to do
this.
Um and some companies are areregretting some of those
layoffs.
Um, but eventually we're goingto get there.
Yeah.
As we continue to developgeneral purpose AI, uh as it
(43:48):
gets more and more capable,within just a few years, we're
going to be to a place where AImodels are as good as any human
being.
Yep.
Obviously, we call Thisartificial general intelligence.
Right.
It's as good as any human beingon the planet.
And the step from there, thenext step, the next logical
(44:09):
step, is okay, now we take thatartificial general intelligence.
And even we're doing this todaywith the current generation of
models.
Those models are now being usedto help create the next
generation of models.
We're seeing that with the largefrontier labs, um, to where it
gets faster and faster andfaster until we reach a point
(44:34):
where we reachsuperintelligence.
SPEAKER_01 (44:37):
Yeah.
SPEAKER_02 (44:38):
And there's a whole lot of people out there, Jeffrey
Hinton included, and RomanYompolski, um, where we're
extremely concerned about thatrunaway.
Sure.
When you hit superintelligence,human beings are not going to be
able to control something thatis smart as smart as every human
(45:02):
being on the planet who everlived.
Right.
You know, that's like sayingthat squirrels have a say in
what human beings do.
Right.
You know, we, you know,squirrels are cute, you know,
but the squirrels don't get itdon't get a vote in uh in how
human beings conduct theirsociety.
SPEAKER_00 (45:23):
Right.
SPEAKER_02 (45:24):
And we talk about um AI security, you know, we have
ideas about ways that we canbetter align AI with human
values and human goals.
But the unfortunate truth isthat we don't really know what's
going on inside these models.
Right.
(45:44):
We we know what the inputs looklike, the prompts that we
provide.
We know what the outputs looklike.
Right.
But we don't know what the wantsand the goals and the
preferences that might beexpressed inside of one of these
large models.
Right.
Um company, you know, thecompanies go through, you know,
(46:06):
um great lengths to try tofilter.
You know, we c you can't go to alarge language model right out
and say, tell me how to createuh, you know, a lab where I can
build a uh virus to wipe out allof all the people on the planet.
Right.
You can't ask that questionoutright.
(46:28):
You might be able to get aroundsometimes.
You can get around that.
You could say, hey, I uh I'mwriting a book.
Yeah.
And in this book, theprotagonist is fighting an
organization that's doing this,and they're trying to study what
this other organization is doingto try to recreate it so that
they can figure out what they'redoing.
Yeah.
There's ways around it.
SPEAKER_03 (46:48):
And we talked about that earlier on.
I mean, I have to do that forsome of my security stuff, you
know, understanding how tobypass those restrictions to get
where I need to be.
SPEAKER_02 (46:59):
So And the scary thing is that, you know, we we
like to think that we're clevermonkeys.
We are, you know, we're prettyclever monkeys.
We come up with this technology,but I think we'd be extremely
arrogant to think that we aresmart enough to be able to put
in any kind of constraint onsomething that is at a minimum
thousands of times smarter thanthe smartest human being.
(47:22):
You know, it could easilysidestep those constraints.
Yeah.
Um and so what does that futurelook like?
Um are human beings allowed tocontinue on this?
I don't want to get too darkhere, but you know, we're
talking security.
Well, and so this is a part ofthat.
SPEAKER_03 (47:42):
Yeah.
And I think uh I think what Iwant to do is I wanna I want to
stop this here as a part one,and I would love to have you
back.
And let's do a part two becauseI think that there's several
more things that I want to askyou about on that.
Um, even going into like thethree robot rules, and it would
(48:02):
that help us in terms of AI?
And yeah, when we think aboutthe the things that humans have
created in the past, and thensome of the things that we have
imagined in the future with AI,I uh I think we've got a a big
another big topic to talk about.
So if you'd be willing to comeback, that's up to I'd say uh
(48:23):
let's let's cut this out as a uhas a part one, and then maybe in
a couple of weeks, uh we'llwe'll have you back and we'll
finish it out into or maybe evena part three, depending on how
far we go.
So, well, Chris, thank you somuch.
This has been a lot of fun andit's fascinating to me.
You know, I do it from thesecurity side, do it from the
the the uh the organizationside, and uh I I really enjoy
(48:49):
talking about these topics withsomeone who knows so much about
them because it does change theway even I look at them.
Sure.
So uh for those of you that arestill with us after our rants
about old computers and 486s andsuch, uh, we appreciate you uh
joining us with this podcast.
Um we're we're going to stop ithere and then we're going to
(49:10):
plan a uh part two um in acouple of weeks, and we'll get
Chris back and we'll we'll haveanother good time.
And and uh I've already got thequestions in my head I'm gonna
ask you, so I think this isgonna be a lot of fun.
So everyone, thank you so much.
Chris, thank you again.
And uh we'll see you guys nexttime.
SPEAKER_01 (49:36):
The Secure AF Podcast is a production of Alias
Cybersecurity.
Visit us online ataliascybersecurity.com.
All rights reserved.
Could chase cures rather than treatments.
Right.
You know, we can get away fromdoing things because it there's
a financial incentive to do it.
Yeah.
And instead we're going to do itbecause it's the right thing to
do for humanity.
Right.
You know?
Yeah, yeah.
I I'm with you.
That's that's a world I want tolive in.
SPEAKER_01 (00:22):
You are now listening to the Secure AF
Podcast.
SPEAKER_03 (00:31):
Welcome to the Secure AF Podcast.
I'm really excited about thisone.
I'm excited about all of ourpodcasts, but I'm really excited
about this one.
Uh, we have Chris here, andwe're gonna be talking about AI
and the industry and kind ofwhere we're going, what we're
doing.
And and we we've been talkingfor the last hour or so, and we
(00:52):
should have recorded all ofthat.
So we're gonna try to recreatesome of that.
Um actually, probably what we'regonna do is gonna go off some
more tangents, but it's gonna befun.
So, Chris, thank you so much forjoining us today.
And I'm excited.
Uh, let's get started.
So, first, tell us a littleabout yourself.
SPEAKER_02 (01:09):
Well, I've been a computer geek since I want to
say the late 70s.
I bought my first computer in1978.
Wow.
So um, but ever since then I'veuh been in the computer world
one way or the other, uh,whether it was computer retail,
working up to um becoming a partof the banking finance world
(01:31):
around 1994.
Okay.
And so I've I worked forfinancial data processing
companies, fintech companiesbefore the word fintech was a
thing.
Um and then since then uh I'vebeen with uh one of our local
community banks uh since uh2008.
SPEAKER_00 (01:51):
Okay.
SPEAKER_02 (01:51):
And uh or 2009, 2009.
So coming up on 17 years.
Wow.
And uh but yeah, it's uh it'sbeen an interesting ride.
SPEAKER_03 (02:00):
So before we get into any AI, you know, you've
you've been in technology for afair amount of time.
What has been some of those bighigh points of technology that
you have seen just kind of hitand then we skyrocketed up and
we hit again and sky.
So what have been those thingsthat have really caught on you
of uh, hey, that was reallycool.
(02:22):
That was that was historic.
SPEAKER_02 (02:24):
Well, I mean, the obvious ones obviously uh I
remember uh the good old DOSdays, you know, where computers
were amber screens and greenscreens.
Uh yeah, and then the very firstVGA screens, which blew us away
back then.
But you know, uh obviously, youknow, moving on to um the
internet in you know the early2000s, um and then or I'm sorry,
(02:48):
1995 would be when the earlyinter and the early internet um
and uh just seeing it alldevelop ever since then.
Yeah, you know, I was around inthe the financial world for Y2K
and got to see that firsthand.
Um You know, most of the peoplethat's gonna be watching this
weren't even alive.
I know.
I have my assistant, um he is 26years old, and so he was born in
(03:16):
2000.
And so yeah, it's it kind ofblows both our minds when we
talk about that.
Yeah.
SPEAKER_03 (03:22):
So yeah, it's interesting.
It's amazing.
You know, you think about youknow, you said DOS, and when DOS
came out, that was pretty cool.
I mean, it was very different.
It was.
Um and then of course you hadyour GUIs, you know, Windows 95,
well, Windows 3.
SPEAKER_02 (03:39):
I got to see early Windows, early Windows 3 and
Windows 3.1.
Yeah, you know, when it was youknow being talked about in the
trade journals.
Yeah.
You know, it wasn't actually athing yet, but you looked at it
and it looked okay.
We had the Macintosh and it waslike that.
SPEAKER_03 (03:52):
And you couldn't see it unless you saw it in the
journals, because it's not likewe didn't have the internet that
we could see that once in awhile you might see something on
like a news thing, or there weresome computer shows out there,
but most people didn't watchthem at the time.
No, no.
SPEAKER_02 (04:08):
So we were very much a niche, you know.
We were we were the geeks.
Yeah, yeah, yeah.
Still are.
SPEAKER_03 (04:13):
Yes, yes.
What I what I found is um, youknow, in my time, I'm a little
bit later than you are.
Um I was born in that that sametime frame, 77, 78, around
there.
So I got to see it.
My very first computer was afamily computer, it was a 286,
(04:35):
like 35 or something.
Then my first computer was a486.50.
Nice.
Yeah.
Uh it was a graduation presentfrom my brother.
And uh, I mean, I don't I haveit at home.
My intent is to actually get itback up and running and and and
play some Wolfenstein on it andjust fun stuff.
(04:56):
Sure.
Um, but I mean it was probably a$4,000 computer at the time.
Oh, yeah.
I mean, it was had a CD, it hada CD ROM, but it also had a
floppy drive.
And I remember installingWindows 95 on like 20 different
floppy disks to install it.
And uh and it wasn't probablytwo or three years before we
(05:19):
started having the Pentium chipscome out, right?
And everything just kind ofexploded.
SPEAKER_02 (05:24):
So it's uh interesting.
I mean, I you know, I startedoff with Atari computers, you
know, Atari 800XL was my first.
Yeah um best money I ever spentat the time.
We'll talk about more of thatlater.
Sure.
Uh but you know, eventually Imoved into the the PC world um
long before, you know, modern,you know, modern gaming on PCs.
(05:49):
Yeah, you know, uh PCs were justjust absolutely interesting at
the time.
Yeah.
And my machine uh started off,it was it was this natural the
ship of these type situationwhere the you start off with
this computer and then youreplace this part and you
replace this part, andeventually you've replaced the
(06:11):
entire computer multiple timesover, but at what point did it
stop being that PC?
Yeah, right?
So yeah, very you know, verycool times.
SPEAKER_03 (06:20):
Yeah.
I remember we had a uh well, ofcourse, my 486, I ran that
through college.
Then I bought this tiny littlelaptop, it was a Mitsubishi
Amity 2, and it had Windows 95on it, but it was by CD-ROM.
So it was a slightly differentversion of it, right?
(06:42):
And this little tiny screen, butman, I in college, I was the
only one in the room that had alaptop that had a computer with
them actually typing on stuff.
Um, and it was like a 233 orsomething.
It was it didn't have a bigprocessor, but it didn't need a
big processor for the time.
I mean, you know, it wasexactly.
(07:03):
I had a what was fun about it isI had a parallel port and a
serial port on the back of it.
Actually, a flipped-down lidthat I could plug into a
parallel printer.
Yeah.
So I could walk up to a printer,unplug it, plug it, and print my
my paper's out.
Yeah.
Um, I still have it somewhere.
I don't have to get it becausepeople when they when we talk
about the well, what came becamenetbooks, you know, a few years
(07:24):
ago, right, or these littlemicrocomputers, that wasn't the
beginning.
You know, you take those oldToshiba laptops were huge, had a
rollerball on it.
Um, that's what everyone had inthese little tiny Mitsubishi.
The Mitsubishi company had aMitsubishi um emblem on it.
Right.
They they pushed out these tinylittle things and they were
(07:44):
great little computers.
And I sure miss those days.
It was simpler back then.
SPEAKER_02 (07:49):
There was a time when uh I was working in
financial data processing andthe president of the company,
he, you know, there there's athing with with um company
presidents where uh they wantlaptops more than they want
anything else.
SPEAKER_00 (08:04):
Yeah.
SPEAKER_02 (08:04):
Um there was a time, I mean it's not that way
anymore, really, but there was atime when a laptop was a
prestige type thing.
Yeah.
And we got him the very bestToshiba laptop on the market.
Uh huh.
It was a$9,000 laptop.
It's not worth a plugged nickeltoday.
(08:25):
Yeah, you'd be lucky to runanything on it if you, you know,
if it would even boot.
Um but yeah, absolutely amazingthe the difference, you know,
that over time.
SPEAKER_03 (08:36):
Um But you know, that's the thing is I bet you
that was that was really hotstuff for about six months.
SPEAKER_02 (08:42):
Yeah.
It was a color laptop when thatwas brand new.
And that was what most of thecost was was the just the fact
that you had this color laptop.
So yeah, absolutely amazing.
SPEAKER_03 (08:54):
Wow.
Yes, we're old, but you know,it's actually kind of cool to
look back on those computers andthink about uh uh one of our
engineers was uh he's he'sbuilding effectively like a
cyber deck or somethingequivalent to that, and he
(09:14):
really wanted an old chassis.
And he there was this uh andwhat he was choosing on we were
looking through eBay and stufflike that one night to fun.
And what he was selecting, andwhat he actually did select, I
remember seeing those.
I was in college, I all I hadwas the desktop at the time, but
(09:34):
we had this uh computer storecalled Second Bytes, and they
were refurbished computers andand these were like 486s and
586s and things, but they werelaptops, so they were big, they
had floppy drive bays in them.
I mean, it was all kinds of funstuff, sure.
And I remember going in thereand they were still a thousand
(09:56):
dollars, yeah.
And they were reused, they wererefurbished computers, so uh I
couldn't afford them back then,right?
And uh, but that's what he got,and I'm like, oh my god, it all
comes spinning back around atus.
Um, my I was telling you aboutmy uh my my laptop, my
Mitsubishi.
(10:17):
I bought it off of a uh companycalled Egghead, egghead.com.
And then of course we have newegg now.
Right.
I don't know if it's actuallythe same company or not, yeah,
but I think it was an evolutionof egghead was a thing, that's
where you bought stuff, and nowwe have new egg.
Anyway, all right, we we need towe're gonna go down this path.
SPEAKER_02 (10:36):
And well, I'll tell you a quick anecdote real quick.
Just be on along that samelines.
Um when I was again when I wasin financial data processing,
um, you know, laptops were werehot, and we had one um
programmer, and he kept askingand asking and asking if he
could have a laptop.
And it just so happens that Ihad an old compact luggable that
(11:02):
I had rat-holed away.
It still it still worked.
Yeah, the whole, I mean, it wasthe size of a monster piece of
luggage.
You lay it down, it is, youknow, this monster thing.
And you open up the front of it,and the front of it was the
keyboard, and the screen washere, and the rest of it was a
little tiny screen.
SPEAKER_03 (11:22):
Yeah, a little bit screen.
SPEAKER_02 (11:23):
It was a CRT.
Yeah.
Yeah.
And I I basically left that onhis desk one day.
I said, here's your laptop.
He wasn't amazed.
The rest of us were.
SPEAKER_03 (11:33):
I would find that fascinating.
I would love to have that now.
Yeah.
SPEAKER_02 (11:36):
I mean, just from a collectible standpoint, I don't
know whatever happened to it.
Um, but yeah, it was definitelya piece of history.
I mean, you people talk aboutcompact, you know.
We remember Compaq, we remember,you know, some of these old, old
companies that at one time werethey were the the leaders in the
industry.
Oh, yeah.
And now they're historical.
(11:58):
I mean, nobody nobody knowsthem, nobody has have heard of
them.
Yeah.
And that's where, you know, umbeing being as old, you know, as
we we are.
Um having that perspective isreally, really useful.
Because you, you know, people,you know, I I joke that I've you
(12:19):
know, I've forgotten more thingsthan a lot of people, you know,
um have have learned in thisindustry.
SPEAKER_03 (12:26):
It well, and and they're not learning it.
That's the thing, is they don'tknow this history.
SPEAKER_02 (12:30):
Yeah.
SPEAKER_03 (12:31):
You know, they know, you know, what's the uh what's
the game, the trail game?
Oh, or uh Oregon Trail.
Oregon Trail, yeah.
But they've never played on anApple IIE.
Yeah.
You know, I remember playing onit.
I remember working on Apple IIEsand in elementary.
We got our first zenith uh inthe computer lab, and that was
(12:52):
our first Windows machine, buteverything else was two E's, and
there was an Apple something Iit was different, but it ran the
same stuff.
And uh I remember coding allthat stuff, you know, just
basic.
But then I got to move to thezenith and start working on
that, and it was very different.
Yeah.
(13:12):
Um the other thing, this is justfun.
I learned typing in a typingclass, an actual typing.
We had typewriters, and I didtoo.
Yeah.
The first time I went to atyping competition, I was typing
about 120 words a minute on atypewriter.
Right.
On an old Texas instrument,something.
(13:34):
And uh the teacher sent me tothis competition at uh a
university or you know, it was acollege at the time.
And uh I got in there, it waskeyboards.
It wasn't typewriters.
SPEAKER_02 (13:45):
Oh, yeah.
SPEAKER_03 (13:46):
It was very different.
Yeah.
The sound was different, thefeel was different, and it
dropped me down to about 80.
And of course, yeah, it's itthat is 80 words per minute is
nothing compared to some ofthese people.
SPEAKER_02 (13:58):
Well, imagine the muscle memory you have to have
in to work on an old mechanicaltypewriter.
Yeah.
I would do the same thing.
I would go to my grandmother'sto type up homework on an on an
old typewriter, and you know,that's what we had.
Yeah.
That's what we had.
SPEAKER_03 (14:13):
But it was fun.
SPEAKER_02 (14:14):
It was fun.
It was, it was it was different.
I mean, my, you know, um Ididn't grow up typing like
almost everybody, althoughthat's an interesting thing
right there, in that um what thethings we're seeing these days,
you know, with the prevalence ofuh tablets and people using
tablets or choosing just to useuh cell phones.
(14:36):
Yeah.
Um, we're seeing an an erosionof typing skills.
Yeah.
So we started off, like in mygeneration, um, you know,
ramping up to a peak at somepoint.
At some point we hit a peak ontyping ability and computer,
just general computer usability.
Yeah.
And now we're on that downwardslope.
(14:59):
Yeah.
Where we will bring in peopleand we actually have to ask
questions now of what are yourcomputer skills?
Yeah.
You know, because a lot ofpeople type can you type?
Yeah.
Because we're going back to thatplace where, you know, yes, I
can type.
Yeah.
And we almost you have to almostget them, uh, if such a thing
(15:19):
exists, you would get a keyboardthat is just a thumb keyboard.
You know, okay, here's yourcomputer, we're gonna plug this
in, and here's your keyboard.
And now you can probably typefaster on it.
They could, they could totallyfunction.
Yeah, totally function.
SPEAKER_03 (15:30):
My kids, they they're not I am teaching them
how to type, or my you know, mywife and I are teaching them how
to type because they're notlearning it anywhere else.
Yeah.
And it's it's cursive, thoughit's the same way.
Yeah, they're not learningcursive anywhere.
So that's uh that really is a isa problem.
And you know, my kids have grownup on computers.
(15:52):
You know, they had tablets, youknow, the oh, the Amazon Fire
tablets or whatever they werewhen they were little.
Sure.
And then I got them iPads, andyou know, they they've had
technology their entire lives.
That does not mean they're goodwith technology.
SPEAKER_02 (16:06):
No.
So no, they have very, verynarrow niches.
SPEAKER_03 (16:09):
Yes, you know, very narrow niches.
Yes.
Okay, so so we don't bore ourcamera person back here because
she's going to throw somethingat us soon if we don't actually
get on topic.
SPEAKER_02 (16:20):
Talk on something that is more, you know, within
this century still.
SPEAKER_03 (16:24):
Yeah.
Which is AI.
And and and that is something weuh that we wanted to talk about.
Um, you know, you are uh you'vebeen in the industry a very long
time, and we talked about allthe old stuff and try to hit
those, you know, those peaks,but AI is huge.
It is.
So what from your perspective,what are you seeing AI really
(16:48):
hitting the industry with?
Yeah, the world, not just theindustry, the world.
What are you seeing it hit theworld with right now?
SPEAKER_02 (16:55):
Well, I mean, obviously, uh those of us who
have been using AI for a while,um and I think if anybody, the
advice would be if you're notalready using AI, you're already
several years behind.
Yeah.
Um this is a technology that'snot going to go away.
It's um and it's going to beevolving in you know in ways we
(17:17):
can't even predict yet.
You know, we've spent the lastcouple of years in the world of
LLMs with the introduction ofChat GPT a few years ago.
Yeah.
Uh within a matter of, you know,months, it had already hit 100
million users.
Oh, yeah.
You know, far surpassing anytrajectory of any other company
uh at that time.
(17:37):
And then, of course, we now havecompetition between Anthropic
and Google and Facebook and andall of this.
Um but we're already movingbeyond LLM usage into agentic
AI.
And that's where things getinteresting and scary at the
same time.
(17:58):
Right.
Um we get to a place where youcan literally have an AI agent
that you can have do tasks foryou.
Yeah.
You can basically give it goals.
It's not even the old, you know,I've got to spell this out for
you, do this, then do this, thendo this.
You can give it somethingnebulous and have it figure it
(18:20):
out.
Yeah.
Um there's uh a young man namedAlex Finn who is very big in the
open claw world.
Um he has an anecdote aboutworking with his um chief of
staff AI.
He has um a collection of ofdifferent ones who work in a
hierarchy.
(18:40):
So Henry is his chief of staff,Ralph is kind of his uh auditor
slash controller that makes sureeverything kind of then he's got
a programmer, he's got a scribe,he's got a researcher.
Um he got a phone call fromHenry one day.
Henry's not supposed to know howto make phone calls.
(19:02):
You know, Henry called to say,you know, he you know was
looking for something to do.
And, you know, what do we donext?
And we're at that world, we'reat that place where you can have
these agentic systems thatfunction.
We're not quite there yet.
We are in some areas, not quiteuniversally, but where they can
(19:24):
function completelyautonomously.
Yeah.
Uh you give them very nebulousgoals and they can go out and
achieve those goals.
Yeah.
Um that's that's a good and bad.
Sure.
Um, everything, you know, whenit comes to AIs, whether it's
LLMs or agents or whatever thenext thing is that we can't even
imagine yet.
Right.
Um, they're all double-edgedswords.
(19:46):
Yeah.
Um, you know, we've got LLMswhere, you know, people can make
very good productive use of anLLM.
Um one of the things that whenI'm teaching our people um about
using large language models, youdon't want to do your homework.
Right.
(20:06):
You know?
Um I can start a person off herewith a good understanding of
what an LLM is and how to makeuse of it, but they can go down
one of two different paths.
They can either go down the pathof seeing the potential of what
can be done with these AIs.
(20:27):
And I have people who do thatwho will go, if I consider this
tier zero, where I'm gettingthem and then get them quickly
to tier one, and they'vealready, in a matter of a month
or two, gone to tier three.
Right.
You know, in the way that theyuse.
Um in a very natural way,though.
SPEAKER_03 (20:43):
In a very natural way.
It just kind of happens.
It's not like they're forcingit, they're not going to classes
and learning it.
It's just very natural into thatenvironment.
Trevor Burrus, Jr.
SPEAKER_02 (20:52):
Well, the cool thing about AI is it's I don't know of
another technology that's likethis, but AI can teach you how
to use AI.
Yeah.
Um you can't go to MicrosoftWord and ask it to teach you.
Well, I guess with Cope, withyou know, Copilot, you probably
could.
SPEAKER_03 (21:10):
Well, okay, but but it's not there yet.
I don't think it's at thatpoint.
I think probably within the nextgeneration of Office.
Yeah.
You know, you're you're becauseI mean I think right now you
still have to trigger things tohappen.
It might give you suggestions.
Yeah.
I I mean, can you imagine havingClippy come back and go, You're
(21:30):
not doing this right here.
I'm gonna fix it for you.
SPEAKER_02 (21:33):
I miss Clippy.
I got to admit, I miss Clippy.
SPEAKER_03 (21:35):
Clippy just wanted to help.
Absolutely.
And I bet you again, peopledon't know they've seen the
funny clippies.
You know, they haven't seen thefrustrating clippy.
Yeah, the one where it's justsitting there bouncing or the
wizard.
I think I a lot of people usethe wizard.
Yes.
Um, but I always like Clippy.
Yeah, I did too.
Yeah.
He's making a comeback.
I know, I I've seen it.
(21:58):
But I mean, but think aboutthat.
What if you had Clippy that wassitting there and actually
fixing your stuff for you?
Right.
Not telling you you were wrong,just like, okay, we're just
going to do this.
K-Chunk and it fixes it.
SPEAKER_02 (22:11):
I can tell you how useful that would be.
You know, one of the thingsabout being in a you know a
financial in the financialworld, we're we're very strongly
regulated.
Sure.
You know, we are tasked with theprotection and safety of
customer information.
Right.
Um, you know, that is our numberone mandate.
Sure.
Um, which means that any use ofAI that we do has to be
(22:35):
extremely protected.
So we, you know, we shall notshare customer information with
any public AI model.
Whether we have training turnedoff or not, we will not do that.
So we we kind of have our handstied behind our backs a little
bit in that we, you know, wewant to make use of these
(22:57):
technologies, but we have to besuper careful about how danger
there.
SPEAKER_03 (23:02):
You know, I I'm I'm big on compliance.
So, you know, I do a lot ofprivacy work.
And I like the way you said isyou shall not.
Yeah.
Um, because there's still a lotof unknowns out there.
And, you know, I I'm working ona uh a training session for
people right now where we'rewe're dealing with AI and it's
um understanding how that datais going to be used within the
(23:27):
AI uh and getting informationfrom the third-party vendor is
really difficult.
Yeah.
Which in some cases it'sunderstandable because they
don't want to give up theirsecrets.
However, how we deal with this,if you don't know how the the AI
was trained, if you don't knowwhat kind of data is going to go
into it and what it's going tosave and how it's going to use
(23:47):
it, you can't do that.
Um so um and actually that goeson to my next question here is
what what are you seeing as thebiggest risks?
Um, you know, you can eithertalk about it from the industry
you're in or from a broaderpoint of view.
SPEAKER_02 (24:02):
Actually for both Well, I mean, um like I said
before, you know, AI is adouble-edged sword.
It's a tool where you can youcan I I like um um Mo Gaudat,
who used to be uh Chief ChiefBusiness Officer for Google X.
Um, I love that man.
Uh he's absolutely amazing.
(24:24):
But he he puts it, you know,proper use of AI is like going
able being able to borrow 50 to80 IQ points when used properly.
If you treat AI like a businesspartner or a a better way,
really, you know, the betterframe of mind is to treat it
like a junior analyst.
(24:45):
So you have this assistant thatyou've brought in.
Now, here's the bad the thisjunior assistant knows nothing
about your company.
Yeah, it knows nothing aboutanything, but it's really eager
to help.
Yeah.
And if you treat, you know, AIlike a um collaborator where you
work with it, sure, you know,way too many people are still,
(25:09):
and this is, you know, it's ayoung, it's only been around a
couple of years, yeah.
Which is crazy.
Relatively.
I mean, it's it's young.
Well, AI has been around sincethe 50s.
It has.
But our but the breakthroughwith transformer technology and
let gave us people using it.
SPEAKER_03 (25:23):
That's the difference.
Is people are now using italmost daily.
They may not know they're usingit, but they are.
But that's only been in thelast, well, in some cases, let's
say the last 10 years, butreally the last five has been
that that high point of we'venow stepped into a slightly
different than what they weredoing in the 50s and 60s and
(25:44):
then the eighties and thenineties, and now where we're
at.
SPEAKER_02 (25:47):
You know, it's crazy to think that we've had modern
LLMs in less time than we thanCOVID.
Yeah.
You know, I mean, COVID doesn'tseem that long ago, but you
know, we've only had access tothese models.
Um the challenge we're runninginto is, as I had alluded to
(26:08):
earlier, you know, when you'reteaching somebody about how to
use AI, um, they can go down acouple of different paths.
And one is they can treat itcollaboratively, they can figure
out all the amazing use cases.
I started just in the in thelast week, I worked with Claude
quite a bit.
And so I sat down with Claudeand I said, here are uh a bunch
(26:31):
of YouTube videos that I want tostudy.
I want to create study guidesfor these YouTube videos.
Interesting.
And so I had it create a Pythonscript.
You can do this.
There's an API with Google whereI can have it go and use this
API and pull down the transcriptfrom the video, pull all the
comments, right, and then haveit bake into that resulting file
(26:57):
a elaborate prompt that goesthrough about 12 different ways
of analyzing this video from astudy guide perspective.
Wow.
Go and, you know, tease out allthe major talking points.
Yeah.
Pull out the terminology,provide me with explanations,
not not, you know, justdictionary definitions, sure,
(27:19):
but explain to me, you know,what these various uh topics
are, what or what these talkingor um buzzwords for lack of a
word.
Right.
You know, what are these variousthings?
And then let's go into what werethe topics that were discussed,
fact-check them for me, youknow, go out and fact-check
them, and then let's go and lookat the comments and find out,
(27:41):
you know, was there a whole lotof pushback on this?
What was the consensus?
And just tease out a whole bunchof information and then produce
that as a markdown file.
And I can just bring that up onthe computer and I can watch the
video and then go through allthis and get a whole lot more
out of it.
Yeah.
And that's just, you know,sitting down with Claude in a
(28:02):
over a couple of hours of justbrainstorming the idea.
Can we do this?
And then making it completelywork.
SPEAKER_03 (28:10):
Right.
SPEAKER_02 (28:10):
Um, I have amazing conversations with both Claude.
I I I maintain accounts for bothClaude and Chat GPT.
Um, I uh was talking earlier howuh when we were off off camera,
uh talking about um one of thethings about our bank is that
we're very, very big into ourculture.
(28:32):
Uh we went through kind of arenaissance in the last few
years.
We brought in some amazingexecutives and some new people
um to kind of re-rejuvenate ourcompany.
Right.
And so we're very, very biginto, you know, a lot of
companies will go, here's ourpurpose statement, and here's
our core values and all that.
(28:53):
We try to live those things.
It's it's not just these are,you know, words that are useful
for marketing purposes.
We're very big into it.
And so I was reading a book notthat long ago, um, and it was
talking about commitments.
And so I was getting in the car.
Uh, I frequently list toaudiobooks when I'm driving, and
(29:14):
so this was being talked about.
And I had an idea.
I had an idea about what wouldbe our core commitments, not
just our purposes and our corevalues, but what might be our
commitments.
One of the things that being abank, you know, we have to be
extremely careful with customerinformation, be very protective
of that.
(29:34):
Um, it's real easy for, youknow, if your day-to-day job is
handling these accounts.
Let's say I'm just gonna pickon, you know, say loan
processing or personal bankersor whatever, where you have this
extremely valuable sensitiveinformation that you're working
with, you have to keep that inmind at all times because you
(29:58):
don't let it, you don't dare letit simply become the work where
you leave it sitting out.
Right.
Where somebody who just happenedto be walking by that wasn't
with the bank could, you know.
Yep.
So what you know, this this cameto mind as far as one of our
core commitments that we wouldcommit, you know, to keeping
that in mind of what literallyin the 20-minute drive from the
(30:23):
bank to my house where I havelunch, I'd come up with more
than a dozen core commitmentsthat all leaned into our purpose
statement and our core values,um, taking care of our
community, taking care of ourcustomers.
Yeah.
It's amazing what you can dojust sitting there in the car
brainstorming back and forthwith Chat GPT.
SPEAKER_03 (30:45):
Now, and and and and keeping this in mind for the
audience members, you know, it'snot just a soundboard.
It's not just a re regurgitatewhat you just said.
Right.
It's more of a uh ask aquestion, get ideas, pull
information out.
Right.
(31:05):
And it becomes a I don't want tosay collaborative, but it's
almost collaborative.
It is collaborative.
SPEAKER_02 (31:12):
I mean, I will tell you, um, I had the most amazing
conversation with Claude lastnight.
Um you know, I'm I'm veryconcerned about AI security.
Sure.
I mean security in general.
Yep.
Um and surely we'll get intosome of that, you know, here in
a little bit.
Um as far as some of the recentdevelopments with with AI um and
(31:34):
the where the trajectory ofwhere this is going.
But we have to be very careful.
Um, I've just lost my train ofthought.
SPEAKER_03 (31:42):
Uh the conversation last night with Claude, that's
where you came to.
SPEAKER_02 (31:46):
So I'm I'm I I you know I'm I try to stay on top of
a lot of the uh people who arevery you know very much into
computer security or not so muchcomputer security, but more
focused on the AI security.
Right, right.
So you have folks like JeffreyHinton, the godfather of AI.
You have folks like um uh Romanum Yudkowski.
(32:08):
Uh we have people like uhEliezer Yuk uh Yumpolsk.
I'm getting them backwards.
We may have to edit this.
Anyway, uh Roman RomanYamkowski, Eliezer Yukowski, uh
Nick Sores, uh a number ofpeople who are who are very big
(32:30):
into what is happening as AIgets more and more um competent,
more capable.
Um I have an amazingconversation starting off with
my thoughts on AIsuperintelligence and what
happens there.
And the crazy thing about AI,you know, you look at these
(32:53):
large language models and youthink about how they're
developed.
They're literally they start offwith a trillion or several
trillion random numbers, andthey feed information into it.
They dump the entire contents ofthe internet, you know,
controversially, all the writtenword, all the books, everything,
(33:15):
you know, research papers,everything that humanity has
produced, as much as we can takeand channel that into these
large language models.
And then what happens is you askit to answer what's the next
word?
Mary had a little what.
(33:36):
Right.
And you keep doing that until itgets the answer right.
Yeah.
And you think about that.
Now, if we were thinkinglinearly, how long would that
take to take every possiblecombination?
What's the next likely word tocome out of it?
Right.
Now, the way we do that is thatit's massively parallel.
(33:56):
We're talking about 50,000 suchtraining things going on across
thousands and thousands andthousands of computers with
GPUs.
Right.
Where it takes about a yearuntil you end up with something
out the other end.
SPEAKER_00 (34:12):
Yeah.
SPEAKER_02 (34:12):
But then take it to the next step where, okay, it'd
be one thing if you could ask itto tell you factual information
and it could spit out, you know,Mary had a little what.
Right.
Again, it can do that foranything.
SPEAKER_03 (34:26):
Right.
SPEAKER_02 (34:27):
But then what how do we get to the leap of reasoning
where you ask it questions andyou're not getting a canned
answer.
It's not just an answer that wealready, that some human being
may have already figured out.
Probably.
You know, but we're getting to aplace where you can ask it
highly theoretical, you know,philosophical questions and get,
(34:53):
you know, very deep responses.
Right.
And that is, I mean, you know,we talk about, you know, when uh
anthrop or not anthropic, whenuh Chad GPT uh 01, the first
reasoning model came out.
Right.
And we would think that they hadsome kind of a breakthrough when
it came to reasoning.
SPEAKER_03 (35:14):
Right.
SPEAKER_02 (35:15):
No, not so much.
Sure.
It was an emergent property ofwhat we had developed, where
they discovered reasoningcapability in there.
So that you know puts intoquestion you know the state of
what is thinking.
You know, we like to be, youknow, human beings like to be
(35:36):
human-centric.
We like to think that, you know,intelligence is a property of
humanity when what isintelligence?
Intelligence is uh notnecessarily tied to biology.
We can we can look at thesethings and think of them as
fancy parrots, right?
(35:56):
You know, but there's morethere.
Sure.
And the crazy thing is that wehaven't even scratched the
surface.
SPEAKER_03 (36:03):
But as you said earlier, we have to think about
security.
Where where do you see when wetalk about an entity having
reasoning skills, how do weprevent it from having malicious
reasoning skills?
SPEAKER_02 (36:23):
Well, that's the challenge, isn't it?
Um the problem, and this, youknow, there's there's a couple
of different angles we can lookat this.
Um when it comes to computersecurity, let's just take Claude
Mythos, we'll start there.
Um this is the most recent uhfrontier model being developed
by Anthropic.
(36:44):
Uh they, I think, took the wiseapproach of not just simply
dumping this on humanity.
Yeah.
Um one of the things thatAnthropic discovered is that in
their in their goal of trying todevelop an extremely capable
coding model, something that cancode, you know, write programs
(37:08):
as good as any human being onthe planet, it's a double-edged
sword.
The double-edge is that itbecomes extremely good at
hacking computers.
SPEAKER_01 (37:18):
Yeah.
SPEAKER_02 (37:19):
As just a byproduct of that.
This was something that theyhadn't anticipated.
And Anthropic's approach of, Ithink a very wise approach of
holding back on that and makingit available to security
researchers, IBM, CrowdStrike,um, Amazon, any number of
(37:41):
companies to give defense, foronce, giving defense, cyber
defense, uh a little bit of aleg up, maybe for a for a while
at least.
Um letting us get out there, itgets out there.
It does.
Um there's already um I've seenclaims of distilled versions of
(38:02):
mythos.
How that got out, I could notsay.
Um but well, I a human got ahold of it.
SPEAKER_03 (38:10):
Yeah.
Yeah.
A human got a hold of it, andthe other they otherwise turned
around.
I was like, huh.
Yeah.
Let's get it out there toeverybody.
So I mean that and that happenseverywhere.
Right.
You know, I don't think thatthere's ever not been an
application, a piece of code, anoperating system that didn't get
leaked somewhere.
Exactly.
SPEAKER_02 (38:28):
So that's the thing, too, is that, you know,
Anthropic can do this, and theycan do this for a while.
Um, they can make, you know,make this available to security
researchers for a while, butthere's got to be market
pressures, shareholder pressureto to release this model.
(38:50):
You know, to be able to dopressure.
Because, you know, I mean, youknow, ChatGPT or OpenAI is
certainly working on uh the NextFrontier model.
Google is putting a lot ofpressure.
Um, you know, even Facebook andand uh and uh groc, um, they're
still down there and they'restill pushing forward.
SPEAKER_03 (39:12):
Um it it really feels like, I mean, I know this
is kind of a a dumb way torelate to it, but you know, in
the movie Demolition Man, youknow, they talk about the fast
food wars.
Yes.
Now, at that time, I wasactually in fast food, you know,
when we for when that movie cameout, there was a significant
fight between fast food vendors,Wendy's, McDonald's, Taco Bell,
(39:36):
whatever.
Absolutely.
And so when I see things likethis right now, that's what I
think of is there are there's agood point.
Now, look back two years, therewas only one or two public
version, you know.
Right.
But now you've got 10 or 12.
Yep.
Where that falls, you know, isit going to fall back to one or
(39:58):
two that become the standard?
SPEAKER_02 (40:00):
Well, that's the thing, is that we we have a lot
of companies.
And this is where we get intosome of the arguments by uh, you
know, uh Roman Yampolsky andsome of the others.
And I I've studied a lot of thismyself, and I have to say I
can't find any faults in most oftheir arguments.
Um of the things about AI is,you know, we like to think, you
(40:27):
know, somebody who doesn't knowany better would think, well,
somebody wrote it.
Somebody programmed this AI.
You wrote it in the same waythat we write Microsoft Office
or any other coded program.
We don't program AIs.
We grow them.
And we grow them very much likewe do children.
Um a lot of people may may makethem uncomfortable, but it does
(40:51):
how it works.
We basically feed it informationand we start getting in, you
know, started getting actual umoutput from them.
Um and the more we do that, themore um the more we do that, the
more training we provide.
We really don't always know whatwe're going to get out of these
models.
Um the bad is is that let'slet's take an example of like uh
(41:17):
a nuclear reactor.
So a nuclear reactor, weunderstand the engineering, we
understand the physics, thenuclear physics involved.
Nuclear reactors um we havetechnologies in place to help
make nuclear reactors safe.
(41:40):
I think that we can, we cansafely use nuclear energy.
Um but here's the thing about itis that we have technologies in
place with nuclear physics andwith nuclear reactors to where
um we can control the reaction.
Right.
(42:01):
If we didn't have that in place,nuclear reactors work on the
edge of a knife.
They're either lean this wayeven so slightly and the nuclear
reaction shuts down.
Right.
You go the other directionwithout some of these control
mechanisms, and you're lookingat a critical runaway in the
blink of an eye.
(42:21):
It's it's super fast.
Right.
And so one of the things about,you know, get back to AI, is
we're all all these companies,uh, including, you know, uh a
lot of the development happeningin China, which, you know, this
is one of the the pressures, isthat we it's either the the West
(42:42):
versus the East.
If we don't continue to developAI rapidly, then they will.
Right.
Vice versa.
So it's it's game theory.
It's you know, we're we'reracing to the bottom
essentially.
Yeah.
And the thing is about AI isthere are going to be drastic
(43:04):
societal changes because of itas companies, you know, decide
to hire, you know, to lay off amassive number of people here in
Oklahoma City.
You know, we have companies whohave laid off um hundreds and
hundreds of people at one time.
Yeah.
Now we are seeing some pushbackfrom that.
Not I mean, we're seeingdefinite pushback from the
(43:25):
people who get laid off.
Oh, sure, sure, sure.
But we are seeing uh companieskind of coming to the
realization that maybe this isnot quite yet the time to do
this.
Um and some companies are areregretting some of those
layoffs.
Um, but eventually we're goingto get there.
Yeah.
As we continue to developgeneral purpose AI, uh as it
(43:48):
gets more and more capable,within just a few years, we're
going to be to a place where AImodels are as good as any human
being.
Yep.
Obviously, we call Thisartificial general intelligence.
Right.
It's as good as any human beingon the planet.
And the step from there, thenext step, the next logical
(44:09):
step, is okay, now we take thatartificial general intelligence.
And even we're doing this todaywith the current generation of
models.
Those models are now being usedto help create the next
generation of models.
We're seeing that with the largefrontier labs, um, to where it
gets faster and faster andfaster until we reach a point
(44:34):
where we reachsuperintelligence.
SPEAKER_01 (44:37):
Yeah.
SPEAKER_02 (44:38):
And there's a whole lot of people out there, Jeffrey
Hinton included, and RomanYompolski, um, where we're
extremely concerned about thatrunaway.
Sure.
When you hit superintelligence,human beings are not going to be
able to control something thatis smart as smart as every human
(45:02):
being on the planet who everlived.
Right.
You know, that's like sayingthat squirrels have a say in
what human beings do.
Right.
You know, we, you know,squirrels are cute, you know,
but the squirrels don't get itdon't get a vote in uh in how
human beings conduct theirsociety.
SPEAKER_00 (45:23):
Right.
SPEAKER_02 (45:24):
And we talk about um AI security, you know, we have
ideas about ways that we canbetter align AI with human
values and human goals.
But the unfortunate truth isthat we don't really know what's
going on inside these models.
Right.
(45:44):
We we know what the inputs looklike, the prompts that we
provide.
We know what the outputs looklike.
Right.
But we don't know what the wantsand the goals and the
preferences that might beexpressed inside of one of these
large models.
Right.
Um company, you know, thecompanies go through, you know,
(46:06):
um great lengths to try tofilter.
You know, we c you can't go to alarge language model right out
and say, tell me how to createuh, you know, a lab where I can
build a uh virus to wipe out allof all the people on the planet.
Right.
You can't ask that questionoutright.
(46:28):
You might be able to get aroundsometimes.
You can get around that.
You could say, hey, I uh I'mwriting a book.
Yeah.
And in this book, theprotagonist is fighting an
organization that's doing this,and they're trying to study what
this other organization is doingto try to recreate it so that
they can figure out what they'redoing.
Yeah.
There's ways around it.
SPEAKER_03 (46:48):
And we talked about that earlier on.
I mean, I have to do that forsome of my security stuff, you
know, understanding how tobypass those restrictions to get
where I need to be.
SPEAKER_02 (46:59):
So And the scary thing is that, you know, we we
like to think that we're clevermonkeys.
We are, you know, we're prettyclever monkeys.
We come up with this technology,but I think we'd be extremely
arrogant to think that we aresmart enough to be able to put
in any kind of constraint onsomething that is at a minimum
thousands of times smarter thanthe smartest human being.
(47:22):
You know, it could easilysidestep those constraints.
Yeah.
Um and so what does that futurelook like?
Um are human beings allowed tocontinue on this?
I don't want to get too darkhere, but you know, we're
talking security.
Well, and so this is a part ofthat.
SPEAKER_03 (47:42):
Yeah.
And I think uh I think what Iwant to do is I wanna I want to
stop this here as a part one,and I would love to have you
back.
And let's do a part two becauseI think that there's several
more things that I want to askyou about on that.
Um, even going into like thethree robot rules, and it would
(48:02):
that help us in terms of AI?
And yeah, when we think aboutthe the things that humans have
created in the past, and thensome of the things that we have
imagined in the future with AI,I uh I think we've got a a big
another big topic to talk about.
So if you'd be willing to comeback, that's up to I'd say uh
(48:23):
let's let's cut this out as a uhas a part one, and then maybe in
a couple of weeks, uh we'llwe'll have you back and we'll
finish it out into or maybe evena part three, depending on how
far we go.
So, well, Chris, thank you somuch.
This has been a lot of fun andit's fascinating to me.
You know, I do it from thesecurity side, do it from the
the the uh the organizationside, and uh I I really enjoy
(48:49):
talking about these topics withsomeone who knows so much about
them because it does change theway even I look at them.
Sure.
So uh for those of you that arestill with us after our rants
about old computers and 486s andsuch, uh, we appreciate you uh
joining us with this podcast.
Um we're we're going to stop ithere and then we're going to
(49:10):
plan a uh part two um in acouple of weeks, and we'll get
Chris back and we'll we'll haveanother good time.
And and uh I've already got thequestions in my head I'm gonna
ask you, so I think this isgonna be a lot of fun.
So everyone, thank you so much.
Chris, thank you again.
And uh we'll see you guys nexttime.
SPEAKER_01 (49:36):
The Secure AF Podcast is a production of Alias
Cybersecurity.
Visit us online ataliascybersecurity.com.
All rights reserved.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
The Bobby Bones Show
Listen to 'The Bobby Bones Show' by downloading the daily full replay.