WEBVTT

00:00:00.099 --> 00:00:03.960
I'm recording this at the end of a week that was
packed

00:00:03.980 --> 00:00:07.540
with AI-related launches. And you're right,

00:00:08.059 --> 00:00:11.640
every week is packed with AI-related launches.

00:00:11.659 --> 00:00:15.580
It can be rather annoying, but this week we had
Google I.O.,

00:00:15.759 --> 00:00:19.659
and therefore Google gave us tons of AI stuff, and
we'll dive into

00:00:19.719 --> 00:00:23.399
that, and we had the Anthropic event

00:00:23.479 --> 00:00:27.299
where they launched their Claude 4 models, and
that was also

00:00:27.459 --> 00:00:31.219
pretty interesting. Also because we can clearly
see that

00:00:31.299 --> 00:00:34.939
these companies are following totally different

00:00:35.139 --> 00:00:39.059
strategies, and we can see that the models
developed by

00:00:39.139 --> 00:00:42.699
these companies focus on totally different things.

00:00:42.700 --> 00:00:46.420
And in general, we might therefore see that we
maybe

00:00:46.500 --> 00:00:50.479
don't get AGI, general intelligence, through

00:00:50.500 --> 00:00:54.479
those LLMs anytime soon. Instead, we might get a
lot of

00:00:54.759 --> 00:00:58.539
specialized products and models, and that's a

00:00:58.559 --> 00:01:02.019
rather interesting future, and in my opinion, the
more

00:01:02.219 --> 00:01:05.999
likely future than having that, uh, all-knowing

00:01:06.099 --> 00:01:09.639
super AI. But we'll see. And I actually wanna
start

00:01:10.079 --> 00:01:13.839
with Anthropic and Claude 4 because when I'm
recording this, it

00:01:13.879 --> 00:01:16.919
was just yesterday that they launched this all.

00:01:17.319 --> 00:01:21.019
And they launched two new models, two new versions

00:01:21.119 --> 00:01:24.799
of the Claude 4 model, so to say, and if you
watched their

00:01:24.859 --> 00:01:28.779
livestream or if you read their announcement blog
post, which of course,

00:01:28.799 --> 00:01:32.719
like many other sources, is linked below this
episode, then you'll quickly

00:01:32.759 --> 00:01:35.759
notice that they're only talking about coding.

00:01:36.099 --> 00:01:39.619
Their model is only about coding, or their

00:01:39.679 --> 00:01:43.419
models are only about coding, and that seems to be

00:01:43.499 --> 00:01:47.419
the niche. It's not really a niche, I guess, it's
one of the

00:01:47.579 --> 00:01:51.419
things these, uh, large language models can do
best, but

00:01:51.439 --> 00:01:54.939
it's their thing they're specializing on clearly.

00:01:55.239 --> 00:01:58.859
And it was kind of obvious, I guess, because with
Claude

00:01:58.899 --> 00:02:02.799
3.5 and 3.7, though that model

00:02:02.799 --> 00:02:06.759
had some problems, but with those older models
that were released a

00:02:06.759 --> 00:02:10.419
couple of weeks or months ago, we could already
see that they were primarily

00:02:10.539 --> 00:02:14.479
strong in generating code. And in general, these
large

00:02:14.519 --> 00:02:18.279
language models are decent at generating

00:02:18.379 --> 00:02:22.279
code, at least up to a certain point, and I'll
also get back

00:02:22.279 --> 00:02:26.259
to that. So it's no surprise that Claude 4 is all
about code and that

00:02:26.299 --> 00:02:29.939
they only mention (laughs) that in their
announcement essentially, and their

00:02:30.099 --> 00:02:33.699
Claude 4 seems to do really well. Now, I

00:02:33.839 --> 00:02:36.799
personally haven't used it a lot. I played around
with it a bit.

00:02:36.819 --> 00:02:40.239
It's available through Cursor and VS Code, for

00:02:40.279 --> 00:02:44.219
example, but not so much in many other tools,

00:02:44.659 --> 00:02:48.259
and we'll see how well it does because I do
remember that

00:02:48.279 --> 00:02:52.219
when Claude 3.7 was announced, it also was super
amazing and

00:02:52.219 --> 00:02:56.079
all these AI influencers on X hyped it up like
they do with every model, of

00:02:56.099 --> 00:02:59.059
course, and then we all found out, well, it's not
that great.

00:02:59.059 --> 00:03:02.899
It kind of does random stuff, it does stuff you
don't want it

00:03:02.919 --> 00:03:06.759
to do, it changes random parts of your code, and
we'll have to

00:03:06.799 --> 00:03:10.559
see if Claude 4 is any better there. But what's
interesting is that

00:03:10.679 --> 00:03:14.259
Anthropic is clearly becoming that code large

00:03:14.319 --> 00:03:18.199
language model company. They want to win the code

00:03:18.339 --> 00:03:22.079
generation game, which is a difficult game to win
because it's not

00:03:22.139 --> 00:03:26.059
like they're the only model creator that wants to
win there, but

00:03:26.079 --> 00:03:29.659
it seems to be their thing. And of course, the
general

00:03:29.719 --> 00:03:33.419
question is what's the end game for large language

00:03:33.519 --> 00:03:36.899
models and code generation? You could of course

00:03:36.979 --> 00:03:40.679
argue that in a year or so, no

00:03:40.799 --> 00:03:44.519
code or almost no code will be written by human
developers, and there

00:03:44.579 --> 00:03:48.119
have been some CEOs that followed that

00:03:48.179 --> 00:03:52.019
reasoning or that shared, uh, that idea, that, uh,
way

00:03:52.099 --> 00:03:56.059
of thinking. I don't agree at all, and I think
it's total

00:03:56.079 --> 00:03:59.999
nonsense to assume that this will be the future

00:03:59.999 --> 00:04:03.639
anytime soon. What's clear, of course, is that
these

00:04:03.719 --> 00:04:07.519
models can be great assistants, and I've said that
for, for months now.

00:04:07.519 --> 00:04:11.499
I've been using these tools, like probably all of
you, for months now

00:04:11.499 --> 00:04:15.279
as assistants, no matter if that's GitHub Copilot
and the GitHub

00:04:15.339 --> 00:04:18.879
Copilot Agent mode in VS Code or if it's Cursor or

00:04:18.879 --> 00:04:22.699
Windsurf or any other extension or IDE.

00:04:23.259 --> 00:04:25.379
These models can be really good at that.

00:04:25.439 --> 00:04:29.299
By the way, not just when you're using them
integrated into an IDE

00:04:29.339 --> 00:04:32.819
or code editor, but also as a partner for having

00:04:32.879 --> 00:04:36.139
discussions and for evaluating different
approaches or

00:04:36.259 --> 00:04:39.959
architectures when you're planning a project, when
you're setting up a more

00:04:40.039 --> 00:04:43.779
complex project. That's what I'm doing a lot, uh,
I'm using Google

00:04:43.859 --> 00:04:47.539
Gemini 2.5 Pro all the time for discussing

00:04:47.619 --> 00:04:51.599
different strategies, different ways of
implementing certain

00:04:51.679 --> 00:04:55.339
features in a Go application I'm currently
building which I'll hopefully launch

00:04:55.379 --> 00:04:58.439
soon. So I'm doing that all the time, and

00:04:59.139 --> 00:05:02.739
that's where they're really, really valuable, and
of course, that will

00:05:03.079 --> 00:05:06.979
stay that way in the future and they will probably
become more valuable there.

00:05:07.399 --> 00:05:11.059
But when it comes to generating all the code and

00:05:11.679 --> 00:05:15.619
building more complex applications and not getting
lost along

00:05:15.659 --> 00:05:19.499
the way, when it comes to these things, sorry,
that's, that's just

00:05:19.579 --> 00:05:22.199
not going to happen anytime soon, that's totally

00:05:23.079 --> 00:05:26.619
nonsense in, in my opinion. But of course, for
building individual

00:05:26.759 --> 00:05:30.659
features with help of AI, they're amazing and
that's going to stay that way and you should use

00:05:30.679 --> 00:05:34.219
them for that, in my opinion, uh, you should
definitely at least give them a try.

00:05:34.219 --> 00:05:37.999
And I've mentioned it before, I got a course on
how to

00:05:38.019 --> 00:05:41.939
efficiently use, uh, AI tools as a developer if
you wanna learn

00:05:41.959 --> 00:05:43.799
more about that, I'll link it below the video.

00:05:43.899 --> 00:05:47.819
I also got some other AI courses which you'll also
find there, but no matter if you take the course
or

00:05:47.819 --> 00:05:51.485
not, give those AI models a shot.... for using
them, uh,

00:05:51.846 --> 00:05:53.525
as, as assistants when writing code.

00:05:53.665 --> 00:05:57.585
And that seems to be the Claude thing, though of
course they

00:05:57.586 --> 00:06:01.385
clearly, uh, have a different goal. They wanna
replace

00:06:01.406 --> 00:06:04.925
developers in the end, I guess, because they
definitely

00:06:05.005 --> 00:06:08.526
wanna, uh, keep on pushing and build models that
are

00:06:08.645 --> 00:06:11.925
capable of building entire projects on their own
or with

00:06:12.225 --> 00:06:15.866
almost no human input, I guess. And we'll see how
that

00:06:15.946 --> 00:06:18.985
goes, uh, and we'll have to see how good Claude 4
is.

00:06:19.005 --> 00:06:22.585
But yeah, here it is and Anthropic is clearly

00:06:23.005 --> 00:06:26.905
the coding large language model company for now
and we'll see

00:06:26.925 --> 00:06:30.645
if that changes again in the future, but that
seems to be their thing.

00:06:31.126 --> 00:06:33.845
Google, on the other hand, is totally different.

00:06:33.966 --> 00:06:37.866
They also had their keynote and lots of other
sessions

00:06:37.885 --> 00:06:41.806
at Google I/O this weekend. It was all about AI,
as

00:06:41.845 --> 00:06:45.345
you might imagine, and they announced lots of

00:06:45.485 --> 00:06:49.265
initiatives, research projects, a new super

00:06:49.286 --> 00:06:52.845
expensive paid plan that gives you early

00:06:52.985 --> 00:06:56.965
access or exclusive access to certain models and
features,

00:06:57.265 --> 00:07:00.965
uh, if you're in the US. Uh, us peasants from

00:07:01.105 --> 00:07:05.025
other countries are not included for now, but if
you're in the US and

00:07:05.025 --> 00:07:08.945
you're willing to pay 250 bucks per month, after

00:07:08.985 --> 00:07:12.745
the first three months where it's a bit cheaper,
uh, then you get all that

00:07:12.805 --> 00:07:15.905
extra access. And clearly Google is

00:07:16.625 --> 00:07:20.285
trying to achieve a more general form of

00:07:20.325 --> 00:07:23.225
intelligence. They're going all in on agents.

00:07:23.225 --> 00:07:26.525
They announced a couple of research projects,
specifically

00:07:26.685 --> 00:07:30.585
Astra and Marina, that are all about giving you,

00:07:30.865 --> 00:07:34.545
uh, agentic features no matter if it's integrated

00:07:34.645 --> 00:07:38.505
into Gemini or if it's on Google devices, like
their

00:07:38.585 --> 00:07:42.425
Google smartphones, the Pixel devices, uh, which
should be capable

00:07:42.465 --> 00:07:46.225
of helping you on all kinds of tasks without a lot
of input from

00:07:46.305 --> 00:07:48.925
you. So they're definitely going all in there.

00:07:49.045 --> 00:07:52.745
They announced View 3, uh, a video generation

00:07:52.785 --> 00:07:56.545
model which looked pretty amazing. Imagen-4, their
image

00:07:56.605 --> 00:08:00.345
generation, uh, model which, uh, should be pretty
good,

00:08:00.385 --> 00:08:04.145
though (laughs) I just tested it before recording
this and it failed to

00:08:04.165 --> 00:08:06.465
create an illustration, but what can you do?

00:08:06.845 --> 00:08:10.805
Uh, maybe it was also the old model. I'm not even
sure which model I'm using there, to be fair.

00:08:11.025 --> 00:08:14.905
But yeah, they announced a lot and I will say the
Google Gemini

00:08:15.085 --> 00:08:18.105
models, the 2.5 models, are pretty amazing.

00:08:18.105 --> 00:08:21.665
I'm using them all the time, as I mentioned
before, for discussing

00:08:21.985 --> 00:08:24.905
how to implement certain features, how to tackle
certain problems.

00:08:24.925 --> 00:08:28.665
And of course also outside of coding, because
unlike Anthropic,

00:08:29.105 --> 00:08:33.085
Google clearly does not have a goal of only
winning the coding

00:08:33.125 --> 00:08:36.965
game, though like all AI companies, it is a lot

00:08:36.965 --> 00:08:40.865
about coding and code assistance and coding
agents, because

00:08:40.905 --> 00:08:44.865
that's like the quick win, right? Large language
models ultimately are

00:08:44.885 --> 00:08:48.405
about generating text. Code is text, so that's
there for a

00:08:48.405 --> 00:08:52.245
natural fit and where a lot of work flows into.

00:08:52.265 --> 00:08:56.005
But it's not just about that. As you can tell by
the

00:08:56.125 --> 00:08:59.625
benchmark numbers Google publishes, it's not all
about

00:08:59.665 --> 00:09:03.325
code. Um, and about... In general, their
messaging.

00:09:03.325 --> 00:09:06.765
It's about building those agents and about helping
you with all kinds of tasks.

00:09:06.765 --> 00:09:10.745
Of course, Google being Google, they're also
trying to

00:09:10.805 --> 00:09:14.385
integrate AI even deeper into their search.

00:09:14.485 --> 00:09:18.405
Ultimately, search may just turn into a

00:09:18.505 --> 00:09:22.305
chat in the end, like an AI chat where you can
still search for

00:09:22.445 --> 00:09:26.365
things, where you can still ask your questions,
but then you'll get

00:09:26.765 --> 00:09:30.525
a, a result, an answer, which of course will
include links to

00:09:30.605 --> 00:09:34.305
sources, but you will not primarily get those
links as

00:09:34.365 --> 00:09:38.325
Google worked in the past and still does to some
extent today.

00:09:38.325 --> 00:09:39.845
Instead you will get these summaries.

00:09:39.845 --> 00:09:43.125
And that of course has interesting implications
for

00:09:43.725 --> 00:09:47.605
people creating content, writing news articles,
blog posts,

00:09:47.645 --> 00:09:51.245
you name it. Because if you're just feeding

00:09:52.105 --> 00:09:55.925
the Google AI or other AIs for

00:09:55.965 --> 00:09:59.505
their summaries and people never visit your
website,

00:10:00.105 --> 00:10:03.945
what's the point of writing blog posts and news
articles in the first

00:10:04.025 --> 00:10:06.445
place, right? So that will be an interesting
future.

00:10:06.865 --> 00:10:10.745
Um, ultimately, we may have AI summaries of
content that was

00:10:10.745 --> 00:10:14.345
also generated by AI and I'm not sure if I'm
liking that future.

00:10:14.345 --> 00:10:18.185
But anyways, that's where we're going, and that is
what Google also

00:10:18.185 --> 00:10:21.845
announced. They're adding more AI integration into

00:10:21.905 --> 00:10:25.865
Google Search. They are adding this AI mode there
where you

00:10:25.925 --> 00:10:29.405
essentially have this chat mode. Um, so an, an

00:10:29.525 --> 00:10:33.325
open, uh, AI chat without the need for a sign-up
and

00:10:33.345 --> 00:10:37.305
anything like that. And it of course makes sense
because they can't just let,

00:10:37.705 --> 00:10:41.665
uh, OpenAI and other providers eat their lunch
with their models, but

00:10:41.685 --> 00:10:45.325
of course it's an interesting challenge for
Google, and

00:10:45.365 --> 00:10:48.885
ultimately as mentioned, also for content
creators.

00:10:49.525 --> 00:10:53.325
Now, of course we could dive into all the details
of

00:10:53.445 --> 00:10:57.125
the different capabilities and projects and models
that were

00:10:57.145 --> 00:11:00.945
announced. You probably already heard a lot about
them, and it'll

00:11:00.985 --> 00:11:04.325
be interesting to see which of these projects

00:11:04.505 --> 00:11:08.465
become reality, anyways, because announcing a

00:11:08.545 --> 00:11:12.465
lot of stuff is one thing. Having good products in
the end

00:11:12.465 --> 00:11:16.265
is a totally different thing, and that's been the
problem with

00:11:16.365 --> 00:11:19.485
AI for the last two years. There's been a lot of
hype.

00:11:19.485 --> 00:11:21.025
There have been a lot of announcements.

00:11:21.105 --> 00:11:25.025
And don't get me wrong, there have been a lot of
great

00:11:25.305 --> 00:11:29.125
products and services and capabilities and we're
really

00:11:29.825 --> 00:11:33.585
living in an amazing or a scary world, depending
on how you look at it,

00:11:33.745 --> 00:11:37.385
when it comes to AI, but there also have been a
lot of

00:11:37.465 --> 00:11:41.285
broken promises. I've heard lots of CEOs tell me

00:11:41.285 --> 00:11:45.265
that, uh, developers will be replaced in six
months, 12 months,

00:11:45.265 --> 00:11:49.105
and that was two years ago. So lots of things

00:11:49.185 --> 00:11:53.025
being said. You should not take it all that

00:11:53.105 --> 00:11:56.985
seriously, but of course it'll be interesting to
see, uh, where we'll be in a

00:11:57.005 --> 00:12:00.165
couple of months from now. And we definitely got

00:12:00.545 --> 00:12:04.225
interesting models, great coding assistants as
mentioned, great

00:12:04.325 --> 00:12:07.245
models for doing research and all that fun stuff.

00:12:07.265 --> 00:12:11.065
And yeah, that's essentially what happened, I
guess not just

00:12:11.205 --> 00:12:14.545
this week but what also kept on evolving this

00:12:14.605 --> 00:12:18.225
weekend. Therefore, please definitely let me also
know what your

00:12:18.325 --> 00:12:21.645
thoughts are about this entire development, about
this week specifically,

00:12:21.925 --> 00:12:25.605
AI in general, and what you think where we'll be

00:12:26.065 --> 00:12:29.945
in, let's say, six months from now. I'll of course
keep you updated

00:12:29.945 --> 00:12:33.725
with my thoughts and so on, and um, wish you a
great time. Bye.