GPT 4o Vs Claude 3.5 Sonnet - Head to Head Comparison - Who wins?
Summary
TLDRIn this comparative video, two leading AI models, Chat GPT 40 and Claude 3.5 Sonet, are evaluated across various writing tasks including blog, academic, and creative writing. The video also assesses their capabilities in file uploading, data visualization, summarizing documents, brainstorming ideas, and interactive visual creation. The conclusion emphasizes the strengths of each model and suggests that combining both can yield optimal results for different tasks.
Takeaways
- 😀 The video compares two top AI models, Chat GPT 40 and Claude 3.5 Sonet, across various categories including writing, academic writing, creative writing, file uploading, information synthesis, summarizing, brainstorming, data visualization, and image creation.
- 📚 In the writing category, both AIs provided similar outputs with basic titles and introductions, but Claude was noted for a more engaging introduction and relevant trends in the content.
- 🎓 For academic writing, Claude was favored for its higher quality academic language, better organization, and critical analysis, while Chat GPT 40 was noted for its strong references and clear argument structure.
- 📚🎨 In creative writing, Claude's output was more natural, fluid, and poetic, introducing characters and setting the scene effectively, outperforming Chat GPT 40's mechanical and basic approach.
- 📁 Regarding file uploading, Chat GPT 40 allows double the number of files (10) compared to Claude (5), making it superior in this aspect.
- 📚🔍 In synthesizing and extracting information from key documents, Claude demonstrated better integration and critical analysis, providing more references and a more connected flow.
- 📚📈 For summarizing a single document, Chat GPT 40 provided a clearer and more structured summary, making it easier to follow and understand the key points.
- 💡 In brainstorming ideas for business names and slogans, Claude initially performed better with business names, while Chat GPT 40 provided more compelling slogans.
- 📊 Both AIs produced similar quality line graphs for data visualization, but Chat GPT 40 offered the option to download the graph, adding convenience.
- 📊🔍 When analyzing a graph or diagram, Chat GPT 40 provided a more comprehensive overview, while Claude offered deeper insights and reflections, suggesting the use of both for a comprehensive analysis.
- 🚀 Claude excels in creating interactive animated visuals, a capability not available in Chat GPT 40, making it the clear winner in this category.
Q & A
What is the main purpose of the video?
-The main purpose of the video is to provide a head-to-head comparison between two top AI models, GPT-40 and Claude 3.5 Sonet, across various categories such as writing, data visualization, and interactive visual creation.
Which three types of writing are explored in the video?
-The three types of writing explored in the video are blog writing, academic writing, and creative writing.
What is the prompt given for blog writing in the video?
-The prompt given for blog writing is to write a 500-word blog post on technological trends to watch out for in 2024.
How does the video compare the AI models' outputs for blog writing?
-The video compares the AI models' outputs by examining the titles, introductions, content, and writing style, focusing on the inclusion of relevant technological trends and the quality of writing.
What criteria are used to evaluate academic writing in the video?
-The criteria used to evaluate academic writing in the video include the quality of writing, the academic tone, the accuracy of references, and the quality of references, such as the use of influential references.
What is the prompt given for creative writing in the video?
-The prompt given for creative writing is to come up with a story based on a world that has ended, where survivors must navigate the aftermath, facing challenges and dangers, seeking safety and a new beginning.
How does the video compare the AI models' outputs for creative writing?
-The video compares the AI models' outputs by examining the natural flow of the story, the use of descriptive language, the introduction of characters, and the overall quality and style of the creative writing.
What limitations are there regarding the number of files that can be uploaded to each AI model for analysis?
-Chat GPT 40 allows uploading up to 10 files at a time, while Claude 3.5 Sonet allows uploading a maximum of 5 attachments to a chat.
How does the video compare the AI models' abilities to synthesize and extract information from key documents?
-The video compares the AI models by uploading the same five documents to each model and asking them to write a literature review. It evaluates the synthesis of information, the use of an academic tone, critical writing, and the inclusion of relevant references.
What is the video's conclusion about the AI models' performance in different categories?
-The video concludes that both AI models have their strengths in different categories, and sometimes it's better to combine both models to get the best output, depending on the specific task or purpose.
How does the video evaluate the AI models' performance in data visualization?
-The video evaluates the AI models' performance in data visualization by asking them to create a line graph from the same data set and comparing the clarity, accuracy, and presentation of the graphs.
What unique capability does Claude 3.5 Sonet have that is highlighted in the video?
-Claude 3.5 Sonet has a unique capability to create interactive animated visuals, which allows users to engage with different ideas and concepts in an interactive manner.
What is the video's verdict on which AI model is better for creating images?
-The video concludes that Chat GPT 40 is the clear winner for creating images, as it has the ability to generate static images, unlike Claude 3.5 Sonet.
Outlines
🤖 AI Writing Comparison: Chat GPT vs. Claude
This paragraph introduces a video comparing two AI models, Chat GPT 40 and Claude 3.5 Sonet, focusing on writing abilities across blog, academic, and creative writing. The comparison includes blog post writing on technological trends for 2024, where both AIs generate similar titles and introductions but differ slightly in content and style. The paragraph also touches on the AIs' handling of academic writing and creative story prompts, suggesting a nuanced difference in their outputs.
📚 Academic Writing and Creative Storytelling Analysis
The second paragraph delves deeper into academic writing, evaluating the AIs' responses to a prompt about graduate employability challenges. Both AIs provide structured academic tones and references, but Claude's response is deemed superior in organization and argument building. The paragraph also contrasts the AIs' approaches to creative writing, where Claude's narrative is more engaging and descriptive compared to Chat GPT's more mechanical style.
📈 Limitations on File Uploads and Information Synthesis
This paragraph discusses the limitations of file uploads in both AI models, noting that Chat GPT allows for 10 files at a time, while Claude is limited to five. The comparison then shifts to the AIs' abilities to synthesize and extract information from key documents. Chat GPT's output is criticized for lacking synthesis and critical analysis, whereas Claude's response is praised for its intelligent processing of information and integration of references.
📊 Data Visualization and Graph Analysis
The fourth paragraph examines the AIs' capabilities in data visualization and graph analysis. Both AIs produce similar line graphs from a given dataset, but Chat GPT is favored for providing a downloadable image file. In graph analysis, Chat GPT offers a comprehensive overview with notable periods, while Claude provides a detailed analysis but makes a critical error in identifying data series, leading to a preference for Chat GPT's accuracy.
🌐 Interactive Visuals and Image Creation
The fifth paragraph highlights the differences in the AIs' abilities to create images and interactive visuals. Chat GPT can generate a basic image from a prompt but cannot create interactive animations. In contrast, Claude excels in creating interactive visual representations, such as a step-by-step guide for starting a new business, making it the clear winner in this category.
🔍 Comprehensive Evaluation and Conclusion
The final paragraph wraps up the video by emphasizing the strengths of both AI models and suggesting that combining their use can yield the best results. It reiterates the importance of identifying the purpose of use to select the appropriate AI tool and hints at the potential of using both Chat GPT and Claude for a more robust output.
Mindmap
Keywords
💡AI models
💡Writing
💡Blog post
💡Academic writing
💡Creative writing
💡Information synthesis
💡Data visualization
💡Interactive visuals
💡Literature review
💡Summarizing
💡Brainstorming
Highlights
A head-to-head comparison of two top AI models, Chat GPT 40 and Claude 3.5 Sonet, is conducted in a video.
The comparison includes different types of writing such as blog, academic, and creative writing.
Both AI models generated similar titles and introductions for blog writing on technological trends in 2024.
Claude's introduction to the blog was preferred for its engaging writing style.
The AI models provided eight technological trends with some overlap and unique additions.
Claude's output was chosen for its conciseness and relevance in writing about 5G and beyond.
Academic writing was evaluated based on quality, tone, accuracy, and references, with Claude showing a higher quality.
Claude's academic writing was structured and organized in a way that was more aligned with academic standards.
In creative writing, Claude's story had a more natural flow and poetic descriptive language.
Chat GPT 40's creative writing was basic and mechanical, lacking the depth of Claude's narrative.
Chat GPT 40 allows uploading double the amount of documents compared to Claude 3.5 Sonet.
Claude 3.5 Sonet excelled in synthesizing and extracting information for a literature review on graduate employability.
Chat GPT 40 provided a clearer and more structured summary of a single document.
Claude was better at initial brainstorming for business names and slogans.
Chat GPT 40 provided more elaborate and longer brainstorming ideas compared to Claude.
Both AI models created similar data visualizations, but Chat GPT 40 offered a downloadable option.
Chat GPT 40 provided a more comprehensive analysis of a graph, while Claude had a critical mistake.
Chat GPT 40 can create images, unlike Claude 3.5 Sonet, which cannot generate static images.
Claude 3.5 Sonet has the unique ability to create interactive animated visuals, unlike Chat GPT 40.
The video suggests combining both models for optimal output based on the purpose of the task.
Transcripts
hello everyone and in today's video I'm
going to be showing you a head-to-head
comparison of the two top AI models chat
GPT 40 and Claude 3.5 Sonet so the first
category we're going to be looking at is
writing and under writing I'm going to
be exploring three different types of
writing blog writing academic writing
and creative writing so let's start off
with blog writing and I'm going to add a
prompt here in both it says write a 500w
blog post on technological Trend to
watch out for in 2024 I'm going to add
that in clae as well as well and then
I'm going to enter
that okay and let's look at comparing
the outputs so if we start from the
beginning just as a general title What
technological Trends to watch out for in
2024 it's just literally taken my prompt
and transformed that into a title and
here in Claud again top Tech Trends to
watch in 2024 so very basic titles for
both we also get very similar
introductions as we progress through
2024 technological landscape continues
to evolve at a rapid pace and then
bringing new trends that promise to
shape the future and here Claude as we
venture further into the 2020s the pace
of technological innovation shows no
signs of slowing down in fact 2024 is
shaping up to be a pivotal year for
several emerging Technologies do like
the way that Claude has written that
introduction apart from the first part
so now let's look at the actual content
we have been given here about eight
different Trends and with Claude it's
exactly the same eight different Trends
as well so if we look for artificial
intelligence we've got Quantum Computing
now metaverse expansion I'd be a bit
reluctant to speak about that is it
really progressing U there's a lot of
talk about it's just died down 5G and
Beyond sustainable technology
biotechnology cyber security Innovations
now let's look at claude's output again
we've got artificial intelligence we've
got extended reality Quantum Computing
sustainable Tech 5G and 6G networks
expand blockchain robotics biotechnology
so let's take a section that they both
include in their output and let's
compare the writing style so 5G and
beyond for example 5G technology is
already transforming the way we connect
but the roll out and enhancement of 5G
networks will continue to be a major
Trend in 2024 with faster speeds and
lower latency 5G will enable more robust
Internet of Things applications from
Smart cities to autonomous vehicles
moreover Research into 6G Technologies
underway promising even more
evolutionary advancements in
connectivity now let's compare the
similar one in um Claude 5G and 6G
networks expand the roll out of 5G
networks will continue to accelerate in
2024 enabling new possibilities in areas
like autonomous vehicles smart cities
and Industrial internet of things at the
same time Research into 6G technology
will ramp up with early discussions
about the potential capabilities of Next
Generation wireless networks they're
both talking about roughly the same
thing and even the examples are almost
the same autonomous vehicles smart
cities and things so in terms of writing
conciseness and quality I would be more
inclined to use uh clae for so you can
see they're both very similar uh both in
terms of the style of writing and the
output I would say that claud's using
trends that are um probably overall more
relevant as we said with chat GPT
there's only just one that i' would be a
bit reluctant to use this blog post but
again like I said they're very similar
in terms of their output using either
one would probably give you similar
quality okay so the next category I'm
going to be looking at is academic
writing and I'm going to add a prompt
here that says write two paragraphs a
total of 300 words on the challenges of
graduate employability using an academic
tone include accurate references and I'm
going to be evaluating both in terms of
a number of criteria so criteria in this
case is going to be the quality of
writing the academic tone the accuracy
of the references that it comes back
with and the quality of references
whether it uses influential references
or
not so let's see what chat GPT has done
with this initial paragraph So first
sentence is straightforward it outlines
the main issue and then it talks about
the misalignment of skills which is one
of the key problems that we are facing
in graduate employability it uses a very
strong reference which is tomance in
2017 to talk about the lack of um
practical experience and soft skills and
it identifies what this slck is and then
it tells us what the Gap is due to and
it cites another influential reference
which is York 2006 and then concludes
the paragraph with of this graduate
employability challenge which is
securing employment now let's compare
the first paragraph that we have
obtained from Claude now the first
sentence says graduate employability has
emerged as a critical concern in
contemporary higher education with
institutions increasingly scrutinized
for their ability to produce Workforce
ready graduates and again it cited
tomance in 2012 so let's start off with
generally how it's introduced the first
sentence notice that they're both almost
exactly the same in their actual
beginning so graduate employability has
emerged as a critical concern we've got
graduate implo ability haser as a
critical concern and this is to note
that we really have to look at how using
AI because you will see the output that
is generated from a lot of generative AI
tools might be almost identical always
make sure that you input your own ideas
as well um with institutions
increasingly scrutinized her ability to
produce Workforce ready graduates and I
like what Claude has done here it's
given us the critical concern okay and
rather than just starting off with the
the the factors uh that are influencing
the landscape it's gone directly to the
point of saying why this is a critical
concern and focusing specifically on the
uh higher education institutions that
are incapable of producing those
graduates and it tells us that this the
challenges associated with this are
multifaceted and it talks about
structural and individual factors and
then it goes on to the macro level um
the rapidly evolving global economy
characterized by technological
advancements and then it uses Jackson
2016 another excellent reference and
this disconnect is further exacerbated
by the increasing competition for jobs
starts the next paragraph it talks about
the individual level and it talks about
how graduates are struggling basically
they've both come back with the essence
of the argument which is the soft skill
argument the uh problem with the the
graduates face and the problem uh
resulting from the advancement in
technology and this resulting in
universities having to change their
curricul now the way it's structured is
a little different and I think Claude
has actually done a better job with this
in terms of structuring an organization
so what will is that it's organized the
problem in the first paragraph in terms
of um the challenges that are faced the
macro level problems that are coming
from technology whereas you can see that
chat GPT has chosen to add that as a
point later on in the second paragraph
and what it's done is that it's
integrated this issue that's arising
from technology with the challenges that
graduates face um and issuing that
problem statement here at the beginning
of the the first paragraph chpt has has
done exactly the same as well it says
graduates face difficulty securing
employment um but I think the way that
Claude has come back with the output is
um a lot more uh intelligent in its
writing it's the quality of the academic
writing is definitely at a higher level
um and the the structuring of the
overall argument and prioritizing
different elements really does uh come
across better in Wood writing and you
can see there are elements of very
critical writing and building on the
arguments as well so if I only had to
use one I'd probably go with Claude I
like the organization structuring of the
arguments that it's put together I like
the quality of the academic language
that it's come back with and differences
are a bit higher quality in
Claude so for our third writing
comparison we're going to be looking at
creative writing and I've added a prompt
here that says the world has Ed
survivors must navigate the aftermath
facing numerous challenges and dangers
they seek safety and a new beginning in
a shattered world come up with a
creative story based on this make it 500
to a thousand words and that same prompt
has been added to
both we start by reading the first
paragraph of each one the world ended on
a quiet Sunday morning no one saw it
coming not the scientists not the
governments on the everyday people who
went about their lives with a casual
confidence that tomorrow would come just
like today so I think already we can see
that with chat GPT this is not the kind
of creative writing that we'd be happy
uh go with it's very mechanical very
basic but tomorrow never came instead
the skies turned dark the earth quaked
and the cities crumbled into dust the
apocalypse was Swift and merciless okay
not a good start for chat GPT for let's
look at Claude so so we have a title to
the short story that we've asked it to
write the last Horizon the world ended
not with a bang but with a series of
wimers climate change resource Wars and
a final devastating pandemic had left
the planet a shadow of its former self
Mira trudged through the ruins of what
was once a thriving Metropolis her boots
crunching on broken glass and debris The
Silence was oppressive broken only by
the occasional gust of wind whistling
through abandoned skyscrapers okay
already I'm sure you can see the
difference
the the writing is more natural it's
more fluid we've got the introduction of
a character straight away in the first
paragraph we've got use of much more
poetic descriptive language and just by
reading that first paragraph I would say
clae Sonet 100% for me in the story so
I've just read through the entire story
from both chat GPT 40 and Claud Sonet
and there's just absolutely no question
about which one wins in this category
and that this this is definitely Claud
son it and just a few things to point
out in written response that's come from
both we'll see with chat GPT 40 that the
way it's presented the output is
extremely basic so for example if I show
you here the introduction of the
different characters um you can see
basic sentences aara had always been
resourceful her father had taught her to
hunt and fish um and then it talks about
Marcus there was Marcus a former
firefighter with a strong sense of Duty
Lena a nurse who had seen more death and
life in the past weeks and and so on
even the way that characters are
introduced is very basic um the
introduction of the paragraphs is very
basic so for example says here one day
and it just reminds you of second grade
third grade writing um and it does that
quite often so we've got here one night
as well um whereas if we compare that
with how Claude comes up with paragraph
introductions for example you can see
things like heart pounding she weaved
through the Labyrinth of ruins so in
terms of style uh language a sentence
structure Claude is definitely the
winner in this
category so our next category of
comparison is going to be looking at the
number of files that chat GPT 40 and
Sonic can upload so I have here a folder
um that has over 15 different files and
what I'm going to do is I'm going to
initially try and upload them all into
chat GPT to see whether it will be able
to um attach them and you can see here
already that it says you may only upload
10 files at a time so in order to test
that as well I'm just going to choose t
top
files okay and you can see that it's now
uploaded the 10 different files that I
have and I can use them for analysis I'm
going to do the same with Claude so I'm
going to start off by trying to attach
all the files that I have in this folder
and again you can see a message that
says you can add the most five
attachments to a chat please select few
attachments so we can clearly see that
in terms of uploading we can only upload
um five documents into CLA Sonet here
and if I upload
these okay and you can see it's now
uploaded five different documents that I
can use for my analysis so the clear
waer in this category would definitely
be chat GPT 40 which just allows you to
upload double the amount that you get
with clae the next category we're going
to be looking at is synthesizing and
extracting information from Key
documents and as we saw with the
previous category we can only upload
five documents in CLA and so in order to
make this a fair comparison what we're
going to do is we're going to upload
five documents to each of chat GPT 40
and to clae 3.5 sonnet and then we're
going to ask the same prompt to see what
it does with extracting information and
synthesizing it so I'm going to start
off here by adding these five documents
to
both and I'm also going to add the
prompt here that says use all the
attached files to write a literature
review made up of five paragraphs on
graduate employability ensure the
writing uses academic tone synthesizes
the information and uses critical
writing add the relevant references and
by the way you can use this not just
academic papers but for any reports
documents work documents um so I'm going
to enter this for both
let's start off with the output that
chat GPT for o has produced you can see
that it's given us titles for the uh
different elements that it's come up
with and we noticed quite a number of
things now first thing it hasn't
formatted the references accurately it's
just given me the titles of the PDFs the
other thing that I noticed here is that
it's actually not synthesized the
information what it's done is that it's
taken each paper separately and then
written an overview of that paper try to
extract the information from it but you
can see for example that here it focuses
on the information from Clark's paper
here it focuses on the information from
Tom ninon and Jackson's paper here it's
focusing on hes paper and so on each one
is addressed separately and we don't
have that overall integration of the
papers the other thing that I noticed um
about how chat GPT has approached this
is that most of the writing is very top
level it does address the main point but
it's not gotten into a detailed indepth
analysis the critical elements are quite
superficial uh for example it would give
you the processual approach talks about
it and then it says this perspective
emphasizes the importance of continuous
personal professional development and
then this approach aligns with the
concept of lifelong learning okay
where's the analysis where's the
critiques of it where's the connections
that it's made with other parts we don't
get this with the at GPT for o synthesis
okay so looking at claude's output we
noticed that it starts off with a
general overview of all the elements in
the paper trying to come up with a
general statement like I mentioned in uh
one of the previous responses you'll
notice that the introductions are very
similar as well graduate employability
has become a major Focus for higher
education institutions and in chat GPT
we had graduate employability has become
a crucial focus in higher education so
again you can see how they uh process
things in a similar way now where clae
actually differs you will see that it's
been very intelligent in the way it's
processed information now it comes up
with the first approach of employability
and gives us the definition that was
suggested by homes um and then not only
that what it does here so we've got the
homes paper here but what it does is
then integrates other Frameworks from
other papers like timon's paper and it
adds that as a supporting evidence to
that and then it brings us some critical
analysis by showing how critics have
argued against this and then it uses a
combination of other papers and it
continues on so in the next paragraph
you can see then it links to the first
paragraph and says an alternative
positional perspective views
employability through the lens of Social
and cultural capital so connecting with
that first paragraph and then again
provides critical elements and it does
this throughout the output that it has
presented and it also gives us the
references of the papers that we have um
included and you will notice it's
provided us with even more references so
we attached five papers but now it's not
only given us the references of the
actual papers that I attached but also
it's extracted information from within
the papers and it's included these as
references as well whereas I didn't get
any reference with chat GPT now before I
finish this section I want to show you
um how the difference in the connections
and the the flow from Claude and compare
that with chat GPT so for example um
this first section talking about the
evolving concept of employability
and then it tells us that individuals
must continually update their skills to
remain employable now we're moving on to
the next section you'd assume that be
some kind of Link or connection between
them but no it's like a complete
Standalone section and it doesn't have
that flow that we got from um cud so in
terms of synthesizing and extracting
information then I would definitely say
I would go with CLA Sonet for this
category
and so for the next category we're going
to be looking at summarizing just one
document so to start off with I'm going
to attach the same document in each of
chat GPT 4 and clae and I'm going to add
a prompt that says summarize the
attached
document you can see that initially it
gives us the authors it gives us
background information about the report
and then it starts off with the key
points it's very clear in its categories
so highlights what the category is and
then gives us some points within each
one and then it does the same for quite
a number of areas and then you can see
that it also adds a section on
implications at the end and then gives
us a full conclusion of the report now
if we look at the output from Claude you
can see that it's also given us uh some
interesting information however if you
don't have a strong background of the
report and you haven't actually uh read
the entire thing you might find it a bit
difficult to understand key points that
have been uh summarized from the report
so start looking at the first point for
example we've got um labor market
challenges Europe and United States
based labor shortages and slowing
productivity growth by 2030 up to 30% of
the current work hours could be
automated driven by Ai and generative Ai
and shifts to labor demand will require
significant occupational Transitions and
then uh Europe meeting up to 12 million
Transitions and US 12 million by 2030
now let's see what clae has come up with
it says the report analyzes how AI
Automation and other Trends will Impact
labor markets in Europe and Us by 2030
by 2030 up to 30% of current work hours
could be automated um accelerated by
generative AI this could require 12
million occupation transitions in both
Europe and the US by 2030 so you can see
that there are definitely similarities
between both I like the way that chat
GPT 40 has broken it down into different
points and given the uh section A
heading it's a lot easier to follow so
you can see here we've got the same data
but the way it's organized in chat GPT
for is a lot stronger it's a lot clearer
it highlights the points uh a lot better
and then you can see the next section is
skills demand and so again we're given
the overall point and then some
information about that point Claude has
given us that same point but because
it's not structured the same way it is
in chat GPT for you're not really
visualizing the whole content of the
report so in my view I would say that
definitely if I want a summary of a full
report and I want to understand the
different sections in element I would be
going for chat GP
4 so for the next category we're going
to be looking at brainstorming ideas and
under brainstorming I'm going to be
looking at B of ideas for slogans and
business names in the first category
brainstorming let's say business ideas
I'm going to add a prompt here that says
brainstorm five business names for a new
social platform that is similar to
Twitter and threads or X and threads so
I'm just going to add that in
I think either one has done a fantastic
job with this and probably have to
regenerate multiple times to get the
output that I want but if I just had to
base it on the initial output that we
see here you can see that Claude has
done a better job it's taken the essence
of the concept and come up with a word
that might work for example pulse Ripple
Nexus whereas chat GPT what it's done is
actually taken some of the words that
we've mentioned so you can already see
that got here tweet sphere it's got
thread wave so it's taking the actual
words and then just added to it so it
probably say that initially clae does a
better job okay let's try and have a
look at what it does with slogans so for
this next prompt we're going to say
provide me five slogans for Creative
campaign for new wireless earphones I'm
going to add the same to
sonnet see what chat GPT 40 has
unleashed the sound embrace the free
them feel the beat Cut the chords tune
in drop out of wires your sound your way
wirelessly liberate your music elevate
your life in terms of son we've got
untangle your sound experience freedom
for your ears whisper directly to your
soul pocket sized concert anytime
anywhere cut the cord amplify your word
I personally like chat GPT 40's output
better I I'd probably be able to work
with some of these and change them
around I think the slogans from um
sonnet really don't reflect the kind of
message that you'd be able to put out
there as a slogan um so again you can
see CLA did better with the first type
and now chpt for o does better with the
second time so I think with this element
of brainstorming I would definitely be
using both to extract ideas I would say
though from experience brainstorming
with uh both Claude and chat GPT 40 is
that I feel that when you want shorter
brainstorming ideas clae tends to do a
lot better when you want more elaborate
longer IDE ideas I usually find that
chat GPT 40 um does a better job with
this so for the next category we're
going to be looking at how both models
do data visualization and I've attached
the same file with a very simple data
set and I'm going to ask them both to
create a line graph from the attached
data
okay so let's look at the output of both
chat GPT 4 and Claude Sonet you can see
that they've come back with very similar
looking line graphs CL son it has here
at the top Olympic medals line graph
chat GPT 40's title more Olympics gold
and silver medals by country and we've
got the countries at the bottom as well
and we've got the same key for both so
what I'd say is that both produce very
good data visualizations why I do like
about chat GPT 40 and I think it would
win this category just based on that
that you have an option here to download
the line graph as a ping file so if I've
downloaded it now I can now come and
open it and I can use it straight
away now the next area of comparison is
going to be in terms of analyzing a
graph or diagram and I'm going to show
you here a chart that I'd like them both
to analyze and let's see the level of
analysis they both come back with so I'm
just going to start off by attaching the
chart here in both and just going to say
analyze the attached chart and provide
the main
insights okay so let's review the output
produced by both models so starting with
chat GPT 40 see the the approach it's
taken is that it first looks at the
general Trends so it gives us the second
industrialization the postwar boom and
underneath that it tells us what's
happening in the United States and
Europe which are the two um main lines
that we're examining in the chart to
specific observations for each country
and then it comes up with
General uh notable periods for us to pay
attention to which is quite interesting
so it's quite comprehensive in the
feedback now if we look at CLA summary
um it also has some distinctive points
that it starts off with so the time span
explains the um time covered it also
gives us the measurement of both axes so
it gives us like an overview of what
we're looking at so that makes it easier
now with the data series I think it's
done a critical mistake here so if you
look it's saying that there are multiple
lines on the chart each representing
different historical periods or economic
factors and it says that the second
industrialization is the black line now
if we look at the chart we can see that
there are two main lines here and one
that represents the US and the other
represents Europe so this section is
completely wrong in terms of
productivity waves I think it's done a
good job here of um outlining the key
areas and we have a comparable part here
from chat GPT as well when it talks
about notable periods I like that it
provides us with recent Trends as well
gives us information in terms of the
volatility and the impact of major
events so you can see there's a bit more
reflection that's coming out of Claude
overall I would say because of the
mistake that we've seen in Sonet I might
be a bit more reluctant I definitely
have to go back and review the work that
it's produced the approaches are very
different if I wanted a top level
overview of what's going on definitely
start off with chat GPT 40 if I'm
looking to draw deeper conclusions to
understand some uh key relationships
between the variables then Cloe is very
useful so this is again an example where
you would benefit from using both GPT 4
and Claude to obtain different types of
information in your
analysis okay so for the next category
of comparisons we're going to be looking
at creating images so I'm just going to
add a here that says create an image of
a spaceship flying to the Moon right on
it space adventure and the same prompt
we're going to enter that in
clae okay and you can see straight away
that with chat GPT we have the ability
to create images the writing is accurate
as well there seems to be something
coming out of the V but overall it's
accurate but with Claude we know that
you can't create General static images
it doesn't have the ability to create
generate edit manipulate or produce
images so here in terms of image
creation chat GPT 4 is the clear
winner so for the next part we're going
to be looking at the model's abilities
to create interactive animated of a
certain concept so I'm just going to be
writing here create an animated
interactive visual of starting a new
business and I'm going to put the same
and
cl okay so as you can see with chat GPT
um it doesn't have the ability to create
an interactive animation so it will just
write down the key steps for us as we
have seen here but with clo and I
mentioned this in a previous video where
I look at the all the features of Cloe
you can click on the link up here it has
an absolutely amazing ability to create
interactive visuals like the one we can
see here so for example this is is
starting a new business we've got step
one idea generation and then if we click
on next it will take us to the market
research and so on and it's just given
us a nice way to play around with
different ideas Concepts present them in
a different way and you can do this with
all kinds of ideas and Concepts so
Claude is the clearcut winner here if
you want to create any animated visuals
and interactive
dashboards so overall we've seen that
these are both amazing models to use the
best thing to do is to identify your
purpose and sometimes it's even better
to combine both models to get the best
output so I hope you found this video
useful and I hope to see you in the next
video
تصفح المزيد من مقاطع الفيديو ذات الصلة
5.0 / 5 (0 votes)