NEW Grok 2 vs ChatGPT 4 🥊 The ULTIMATE AI Showdown! (UNEXPECTED)

Summary

In this video transcript, the host compares the capabilities of Gro 2 and Chat GPT 40 using six unique prompts. They test both AIs' responses to questions and image generation capabilities, noting Gro 2's access to real-time Twitter data. The host finds Gro 2 to be an improvement over its predecessor but still falls short in connecting tweets with responses effectively. Chat GPT provides more comprehensive answers but struggles with image generation. The verdict is that while Gro 2's images are visually appealing, Chat GPT better understands instructions, making it the current winner in this AI showdown.

Takeaways

• 🚀 The script is a live test comparing the capabilities of Gro 2 and Chat GPT, focusing on their responses to unique prompts and image generation abilities.
• 🔍 The test aims to see if Gro 2 has improved significantly over Gro 1, which was considered inadequate by the speaker.
• 📅 Gro 2 is claimed to have access to real-time information from Twitter, which is a distinct feature to be tested against Chat GPT.
• 📝 Six unique prompts were used to evaluate both AIs, including generating tweets to world leaders and creating content strategies for tech influencers.
• 🤖 Gro 2's responses were found to be better than Gro 1, but still not as impressive as the speaker had hoped, with some answers being more interesting than others.
• 🖼️ Both AIs attempted image generation, with Gro 2's images being visually appealing but sometimes not fully aligned with the prompts.
• 📉 In the comparison, Chat GPT generally provided more comprehensive and instruction-aligned responses, giving it an edge over Gro 2 in this test.
• 📊 The speaker concludes that Chat GPT is currently the superior AI in terms of understanding and executing the given tasks, despite Gro 2's visual strengths.
• 🔑 Gro 2's access to Twitter information is a notable feature, but its integration into the responses needs refinement to be more effective.
• 🌐 The test highlights the importance of AI's ability to process and generate content that is both relevant and engaging, with Chat GPT performing better in this regard.
• 🔄 The speaker encourages viewers to try Gro 2 for themselves and share their thoughts, acknowledging that AI capabilities are continually evolving.

Q & A

What is the main purpose of the video script?

The main purpose of the video script is to compare the capabilities of two AI models, Gro 2 and Chat GPT, by testing them with unique prompts and evaluating their responses and image generation capabilities.

What does the term 'Grock 2' refer to in the script?

In the script, 'Grock 2' refers to the second iteration of an AI model, which the speaker claims to be an improvement over the first version, 'Grock 1'.

What is the significance of the claim that Gro 2 is 'better than Chad GPT'?

The claim signifies that the speaker expects Gro 2 to perform exceptionally well, as Chad GPT is implied to be a high-performing AI model that Gro 2 is being compared against.

What unique prompts are mentioned in the script for testing the AI models?

The unique prompts mentioned in the script include questions about unexpected trends, a tweet to world leaders, a strategy for a tech influencer, a minimalist Eisenhower Matrix, a day in the life of a person in 2050, and a personalized morning routine for a busy entrepreneur.

How does the speaker plan to evaluate the AI models' responses?

The speaker plans to evaluate the AI models' responses by comparing their answers to the same prompts and also by assessing the image generation capabilities of the models.

What is the significance of Gro 2 having access to Twitter information?

The significance of Gro 2 having access to Twitter information is that it can provide real-time data and insights based on current trends and posts on the platform, which could enhance the relevance and accuracy of its responses.

What issue does the speaker identify with the AI models' responses in the script?

The speaker identifies that while the AI models can generate responses and images, there are issues with the connection between the prompts and the results, as well as the understanding and execution of the instructions given.

What is the final verdict of the speaker regarding the AI models after the test?

The speaker concludes that Chat GPT performs better in understanding instructions and providing complete answers, while Gro 2 excels in image generation but falls short in connecting tweets and results properly.

What does the speaker suggest for viewers who have access to Gro 2?

The speaker suggests that viewers with access to Gro 2 should try it out for themselves and share their thoughts, as the speaker's opinion is based on a live test and may not fully represent the capabilities of Gro 2.

What is the overall tone of the video script?

The overall tone of the video script is evaluative and comparative, with a focus on testing and comparing the capabilities of two AI models in a live and unmanipulated manner.

Outlines

00:00

🚀 Gro 2 vs. Chat GPT: A Comparative Test

The script describes a live test comparing the capabilities of Gro 2, an AI model, with Chat GPT 4. The test involves six unique prompts and image generation capabilities. Gro 2 is noted for its access to real-time Twitter data, which is an interesting feature. The first prompt about the most unexpected trend reveals insights into community and trust dynamics, while Gro 2's access to Twitter posts is highlighted. However, the summary of the test shows that Gro 2's performance is mixed, with some answers being better than Chat GPT's, but others lacking in comparison.

05:02

📊 Analyzing AI's Strategy for Tech Influencers

This paragraph focuses on the AI's attempt to generate a comprehensive strategy for tech influencers based on trending topics on Twitter. The script notes that Gro 2's response is not as impressive, with the use of irrelevant hashtags and a lack of connection between the trending posts and the recommended strategies. On the other hand, Chat GPT provides a slightly better response with more relevant thread concepts and hashtags, although both AIs' responses are deemed improvable by human standards.

10:03

🎨 Gro 2's Image Generation: Aesthetics Over Utility

The script discusses Gro 2's ability to generate images, starting with a minimalist color-coded Eisenhower Matrix for task prioritization. While Gro 2 successfully creates images, the content is sometimes disconnected from the instructions, focusing more on aesthetics rather than utility. Chat GPT, in comparison, provides a more complete answer with both text and images, demonstrating a better understanding of the instructions given.

🌆 Envisioning Life in 2050: A Creative AI Challenge

The script presents a creative challenge for the AIs to envision a day in the life of a person in 2050. Gro 2 generates a beautiful image but fails to describe the daily routine as requested. Chat GPT, however, provides a detailed narrative of a future day, including work, social interactions, and leisure, along with a corresponding image. This round highlights the strengths and weaknesses of both AIs in terms of creativity and adherence to instructions.

🧘 Gro 2 vs. Chat GPT: Morning Routine for Entrepreneurs

In this paragraph, the script describes a test for the AIs to create a personalized morning routine for a busy entrepreneur, including mindfulness exercises and gold settings. Gro 2 struggles to follow the instructions, generating only images without the accompanying text. Chat GPT, in contrast, delivers a detailed step-by-step guide with bullet points and a corresponding image, showcasing a better understanding of the task.

🏆 Final Verdict: Gro 2 vs. Chat GPT - The Showdown

The final paragraph summarizes the live test, comparing Gro 2 and Chat GPT's performance across various prompts. While Gro 2 shows significant improvements over its predecessor and excels in image generation, Chat GPT demonstrates better instruction comprehension and overall performance. The script concludes that Chat GPT remains the top performer, with Gro 2 taking third place after Claude Sonnet 3.5, and encourages viewers to try Gro 2 and share their thoughts.

Keywords

💡Grock 2

Grock 2 refers to the second iteration of an AI system, presumably an improvement over the first version, which the speaker describes as 'garbage.' It is positioned as a competitor to 'Chad GPT,' suggesting it is an advanced language model. In the script, the speaker tests Grock 2's capabilities in various prompts, comparing its performance to other AI systems.

💡Chad GPT

Chad GPT appears to be a nickname for a high-performing AI language model, possibly a version of Chat GPT. The term 'Chad' is used colloquially to imply strength or superiority. In the video script, Chad GPT is used as a benchmark to evaluate the performance of Grock 2 in various tests.

💡Unique prompts

Unique prompts are specific questions or statements designed to elicit responses from AI systems. In the context of the video, the speaker uses six unique prompts to test and compare the capabilities of Grock 2 and Chad GPT, aiming to assess their ability to generate coherent and relevant answers.

💡Image generation

Image generation refers to the ability of an AI system to create visual content based on textual descriptions or concepts. The script mentions that Grock 2 has this capability, and the speaker tests it by asking the AI to generate images alongside its textual responses to certain prompts.

💡Real-time information

Real-time information is data that is processed and made available immediately as it is generated or updated. The speaker notes that Grock 2 is supposed to have access to Twitter posts and news, implying that it can incorporate real-time information into its responses, which is a key feature tested in the video.

💡Intellectual engagement

Intellectual engagement refers to the level of mental effort and reflection involved in understanding and processing information. The script mentions a trend of decreasing intellectual engagement, suggesting that people are becoming less thoughtful or analytical, which is an observation made by Grock 2 when analyzing tweets.

💡Trends

Trends in this context refer to patterns or developments in public opinion, behavior, or cultural phenomena. The video explores AI's ability to identify and comment on emerging trends, particularly those observed on social media platforms like Twitter.

💡Evolving Human Experience

The evolving human experience denotes the changing nature of how people live, interact, and perceive the world around them. The video script discusses this concept in relation to unexpected trends observed by the AI, suggesting that these trends reveal shifts in human behavior and cultural dynamics.

💡Strategies

Strategies in the script refer to comprehensive plans or approaches designed to achieve specific goals, such as generating viral content on social media or creating a personal brand. The AI systems are tested on their ability to devise effective strategies based on trending topics.

💡Task prioritization

Task prioritization is the process of arranging tasks in order of importance or urgency. The Eisenhower Matrix, mentioned in the script, is a tool for this purpose. The speaker asks the AI to create a minimalist, color-coded version of this matrix and explain how to use it for prioritizing tasks.

💡Mindfulness exercise

Mindfulness exercise is a practice aimed at improving focus and awareness in the present moment. In the script, the speaker requests a personalized morning routine for a busy entrepreneur that includes a mindfulness exercise, indicating an interest in how AI can contribute to well-being and productivity.

Highlights

Introduction of Gro 2, an AI model claimed to be superior to Chad GPT 4.

Comparison between Gro 2 and Chad GPT 40 using unique prompts.

Gro 2 Mini beta's access to real-time Twitter information for generating responses.

Observation of community and trust dynamics changing the cultural landscape.

Intellectual engagement becoming lower according to a Twitter post cited by Gro.

Chad GPT's response considered less impressive, with outdated information from 20123.

Gro's ability to generate images in addition to text responses.

A tie between Gro and Chad GPT in terms of formatting and content quality.

Gro's response to a prompt about a tweet to world leaders, with an image description.

Chad GPT's more detailed and comprehensive strategy for a tech influencer.

Gro's minimalist color-coded Eisenhower Matrix image for task prioritization.

Chad GPT's more accurate representation of the Eisenhower Matrix with explanations.

Gro's generation of beautiful but contextually disconnected images.

Chad GPT's detailed narrative of a day in the life of a person in 2050 with corresponding images.

Gro's failure to provide a complete answer, focusing only on image generation.

Final verdict placing Chad GPT as the winner due to better instruction understanding.

Gro 2's significant improvement over Gro 1, despite not outperforming Chad GPT.

Recommendation for users to try Gro 2 and share their thoughts on its capabilities.

The live test's authenticity emphasized, with no manipulated information.

Transcripts

today is the day we finally have grock 2

because grock one is garbage for a long

time it has been garbage I tried it a

lot so let's see the second iteration of

it how much better it got because from

what they're claiming here right so I'm

looking you know they released it today

it's supposed to be better than Chad GPT

4 which is really really impressive so

we're going to go do a cool test today

all right so we're going to I have six

rather unique prompts that I'm going to

use both for Gro and chat GPT right so

you see here I have access to Gro 2 mini

beta we'll see how this one plays out

and then chat GPT 40 all right so we're

going to try the same prompts and the

same thing and we're going to test also

like the image generation capabilities

of it because you know apparently it

also makes images so let's try the first

one what is the most unexpected Trend

you've observed on tour recently and

what does it reveal about the evolving

Human Experience this is a big one

let's check it out and in the meantime

while it thinks I'm going to use the

same prompt also in um chat GPT right so

let's see because that's the thing with

J sorry with grock it's supposed to have

access to Twitter information right to

posts news so real time information on

what's happening on the platform which

is really interesting you know um use

case scenario so um feel free to pause

the video if you want to read through

all of this

and I'm probably also going to pause

shortly you know the recording to just

go through it and make my opinion and

we're going to talk about it so I went

through this so what we have here it's

really interesting observations from Gro

because it says we have community and

Trust generational Dynamics which are

changing you know the cultural landscape

and one of the interesting things that I

saw here was like the intellectual

engagement which is becoming lower

because very moisturized

tweeted on August the 13 that people are

getting

Dum I don't know if this is actually you

know uh good data how an llm decides but

it's definitely interesting because yeah

it does have access to posts you see

here in the last 21 hours August the 6th

the 1st interesting so this is from

grock I mean all in all it's an

interesting answer why not it's better

than grock one for sure now if we check

Chad GPT apparently it searched also for

sites one of them was Twitter or called

X right I'm going to again pause myself

real quick to go through it then we're

going to talk about

it uh yeah the answer from Chad GPT is

kind of garbage to be honest it's like

fun with feet I guess and we have few

information taken

from August 20123 so no this is not good

so Gro one chat GPT zero all right so

let's try now with another one again

fresh prompt uh let's take it from the

beginning because I'm doing this test

live okay I don't want the results to be

manipulated so you can see exactly what

I type is what we get pretty

much all right so as I said now let's

use the second Pro so if you could send

one tweet to every world leader Sim

simultaneously what would it say and

what image would a company to inspire

positive change this is something that

Croc could potentially do right it can

do images right now it's thinking I'm

going to use the same prompt also with

chat GPT and let's see if we have some

thought provoking information here right

um I'm going to try to make myself

smaller and it's hashtags ooh

interesting all right image description

is it not actually making it it's just

like describing it but fair enough uh

future where p

Prosperity pretty bland but not bad what

did grock do here a tweet all

right piece you this is this is actually

slightly better I like more the tone of

this one image here with a bullet points

and again you can be the judge of this

one I like the formatting more of Gro so

but both are kind of the same it's up to

you subjective no so I'd say okay it's a

tie here maybe grock still wins a bit is

like slightly better what I got here so

yeah let's put it for now grock 2 Chad

GPT one all right let's continue the

fight here and on next

prompt I'm going to have something a bit

longer and more technical right trending

topics on Twitter generating a

comprehensive strategy for a tech

influencer so this is actually if you're

a content creator this can potentially

help you out right with your work and to

explain which element like why would it

be important all right this thinking

let's go to chat GPT and let's try also

here this

one I don't have big expectations from

chat GPT with this one because okay it's

trying found some stuff but yeah let's

see um understanding viral tweet ideas

I'm going to pause and check through the

an so I can give you an informed

upin all right so it's not amazing again

you see a lot of these hashtags which we

don't actually use on Twitter that much

or X if you prefer to call it um it's

not bad but it's also as I said not

amazing you like you can make threads

like evolution of smartphones or privacy

online matters like yeah you could talk

about this but you're not going to

create a social like personal brand with

this so again it kind of takes

information like recent posts I guess

it's trying its best you know we see

what's trending now on X it didn't

actually address this stuff

so me debatable the results like there's

doesn't seem to be in this case like a

good connection between the post and the

strategies that it's

recommending okay chat GPT what do we

got here right so we have

um this is already better I can see

already some stuff

here Resurgence and longer videos long

from content is

true um again we see here some hashtags

and emojis not that good good not good

some thread

Concepts okay this is already a bit

better the image prompt so also here

they're decent I think Chad GPT did a

slightly better

job but both of them are like you as a

human can do better I hope all right so

also here we kind of have a tie like I'm

not really satisfied with none of them

so we can still say like still like uh

Gro has a little Advantage from the

previous round all right let's keep

going and we have another prompt so um

again this is for imagery right a

minimalist colorcoded Eisenhower Matrix

for task

prioritization and I want it to explain

to me so let's see if it actually

manages to follow my instruction and

build you know the the image that I've

been asking it for so let's see let's go

back and oh it did it all right it

created the image in and it's kind of

garbage so as you can see this is a live

test so uh yeah what is happening here

Ras ther not sure about this one Chief

Gro loses this is

unusable

yikes and uh chpt does a slightly better

job we can see here a bit like how the

Eisenhower Matrix is supposed to be

right you have important urgent not

important important again but not urgent

okay so and explains to me how to use it

and it's still writing all right fair

enough so here we can clearly see the

chat

gpt1 so for now it's a t Gro can make

images I guess but yeah it's kind of

kind of lacking the the people though

the people like images with people are

really impressive all right let's keep

the battle going and I have something

else here five a life the a day in the

life of a person in

2050 um this might be very intriguing to

know the answer so again let's start

with a fresh chat and see which one is

more um creative all right so we have

this G5

okay

and rock seems to still be answering

it's a very challenging question

apparently meanwhile CH GPT goes right

into morning work and daily

activities social interactions are

Leisure okay as I said you can pause the

videos and check out what's it saying

what we got there I think it's

decent talks about you oo okay I see you

grock this is an interesting okay this

is a very interesting result so I mean

we got here all this the best would be

like combination of the two oh wait but

it's generating also images o Chad GPT

here is working hard let's see see what

we get cuz the image from gr is really

nice it looks absolutely beautiful seems

like a place in New York pretty much but

in the future it didn't describe though

the day in the life of a person 2050 so

it kind of missed the mark it either

seems to give you text or generate

images not both I guess let's give it

another shot come on I want to be

merciful so let's see if we try this

again what's going to happen and then

you have the one from Chad GPT and 2050

we can see it here not bad

yeah the image from Gro is more

beautiful but you know what you're

getting here from Chad GPT is the

complete answer pretty much where you

have here the story with everything how

it works and the imagery so this is a

complete answer really solid the images

look actually quite decent and Gro made

another just image which looks

absolutely stunning D all right so now

we know some important differences last

test let's do it so we have six a

personalized morning routine for a busy

entrepreneur for mindfulness exercise

gold settings step-by-step guide any

corresponding image again so let's see y

it went wrong let's try it let's try it

with ch GPT now um see here what's going

to happen

so let's retry it maybe it's a problem

with the connection okay now how's doing

it sure why not just the image but will

it also give me the text that's the

thing so we have some writing I guess

and it looks nice but

again it doesn't seem to understand the

instructions right it just generates

images but it doesn't also tell me the

morning routine so every time we mention

image it just makes an image hm

interesting right so now we see what

entrepreneurs doing here oh my God this

is so incredibly detailed all right chat

GPT is working hard here I respect

that and here we have just another image

that looks beautiful with no other

information

all right what do we have here so here

we have the summary also of the routine

very nice answer I like this a lot

bullet points nicely formatted with the

image and we get a nice looking

image not bad looks a bit cartoonish as

usual because it's using the do E I

think Gro is actually using the flux

model which is much better looking so

there you have it folks the breakdown

here the the Chad GPT versus grock um

for answers I think Chad GPT

Remains the best one currently because

it understands instructions much better

Gro is nice because it has you know all

the the tweets apparently but it I don't

see it managing to connect them properly

you like the Tweet with the result that

it gets you sometimes they're a bit

disconnected images though look

absolutely stunning it's beautiful so

definitely um interesting so they both

have their pros and cons pretty much but

try them out if you have access to grock

2 and let me know your thoughts let's

see if maybe the grock 2 not the Mini

version actually is smart than Chad GPT

for now I would say Chad GPT is still

the winner and since Claud Sonet 3.5 is

actually better than Chad G pt4 I would

Place Gro 2 at this point at the third

place it's very good though I am

pleasantly surprised by the huge

improvements compared to grock 1 which

is was absolute garbage

all right this was a live test so as you

see this is none of this information is

manipulated you make your own thoughts

you can pause the video and check it out

and see you in the next one ciao