GPT4o: 11 STUNNING Use Cases and Full Breakdown

Summary

The video script discusses the recent release of GPT 40, an advanced AI model with impressive capabilities. It highlights the model's flirty voice, its ability to adjust tone based on context, and its potential for real-world applications. Examples include guessing scenarios, interacting with other AIs, tutoring in math, summarizing meetings, and providing real-time translations. The script also explores the model's use in accessibility, such as assisting the visually impaired, and in customer service, where it could handle calls on behalf of users. The potential for abuse is acknowledged, but the transformative impact on various sectors is clear, with the voice aspect of GPT 40 being a particularly exciting feature.

Takeaways

• 🚀 GPT 40 has been announced with some parts already released, and it has voice capabilities that are yet to be released, which are considered exciting.
• 🎥 The model can interact with the world through audio, vision, and text, as demonstrated in a video where an AI guesses what's happening in a recording setup.
• 🗣️ GPT 40's voice has been described as flirty and can be adjusted through system prompts, with the ability to interpret and react to user's requests appropriately.
• 🎤 Two AIs can interact with each other, as shown in a demo where they sing together, showcasing the model's ability to understand and respond in real-time.
• 🕹️ The model can play games like rock-paper-scissors and distinguish between multiple people and voices, indicating its advanced recognition and interaction capabilities.
• 📚 GPT 40 can be used for educational purposes, such as tutoring in math, by guiding students through problems without giving away the answers.
• 📝 The AI can take part in meetings, understand the context, and summarize discussions, assigning names to voices and understanding their preferences or points.
• 🌐 Real-time translation is another capability of GPT 40, as it can translate conversations between English and Spanish in real-time.
• 🦆 The model can provide assistance to the visually impaired, describing surroundings and actions, thanks to its low-latency and high-context understanding.
• 🤖 GPT 40 can be used in customer service, potentially handling calls and interactions with service agents on behalf of users.
• 🎨 The model has explorative capabilities in various fields such as creating caricatures from photos, summarizing lectures, and generating 3D object synthesis.

Q & A

What is the main focus of the video script discussing GPT 40?

The main focus of the video script is to delve into the details of the GPT 40 model, its capabilities, and to showcase various real-world use cases that demonstrate its potential applications.

What aspect of GPT 40 is highlighted as particularly exciting in the script?

The voice aspect of GPT 40 is highlighted as particularly exciting, as it allows the model to interact with users in a more natural and conversational manner.

How does the script describe the voice capabilities of GPT 40?

The script describes the voice capabilities of GPT 40 as being able to interpret user prompts and adjust its tone and style accordingly, such as being flirty, whispering, or even sarcastic.

What is an example of a real-world use case for GPT 40 mentioned in the script?

One example mentioned is the use of GPT 40 for tutoring, where it helps a student understand a math problem by asking questions and guiding them to the solution.

How does GPT 40 handle the task of summarizing a meeting in the script?

GPT 40 is able to listen to the discussion, assign voices to specific participants, understand their points, and then provide a summary of the meeting, including the main arguments and opinions expressed.

What is the potential application of GPT 40 in customer service as described in the script?

The script suggests that GPT 40 could be used to handle customer service calls on behalf of users, negotiating or resolving issues without the user needing to be present on the call.

How does the script address the potential for GPT 40 to be used in scams or unethical ways?

The script acknowledges the potential for misuse but suggests that Open AI is likely implementing guardrails to prevent scammers from exploiting the technology. It also mentions the importance of how information is used, which depends on the users.

What is the role of GPT 40 in the example of real-time translation provided in the script?

In the real-time translation example, GPT 40 acts as an interpreter, translating spoken English into Spanish and vice versa, facilitating communication between two people who speak different languages.

How does the script illustrate the capability of GPT 40 in understanding and responding to visual cues?

The script shows GPT 40 being able to see the world through a camera, describe scenes, and even create a caricature based on a photo, demonstrating its ability to process and respond to visual information.

What is the potential impact of GPT 40's capabilities on accessibility for people with disabilities, as mentioned in the script?

The script suggests that GPT 40 could significantly improve accessibility for people with disabilities, such as by providing real-time assistance for visually impaired individuals to navigate their environment.

Outlines

00:00

🤖 GPT 40's Real-World Applications and Voice Capabilities

The script begins with an introduction to GPT 40, highlighting its recent announcement and partial release. The focus is on the model's voice capabilities, which are yet to be released but are highly anticipated. The speaker discusses the model's ability to interact with the world through audio, vision, and text, showcasing its potential through examples provided by Open AI. The first example involves an Open AI employee using GPT 40's vision and voice to guess the situation in a video, demonstrating the model's conversational and interpretative skills. The flirty nature of GPT 40's voice is noted, along with its ability to adjust based on user prompts. The script also humorously references a 'California Valley Girl' accent, adding a personal touch to the AI's voice.

05:01

🎤 AI Interactions: Seeing, Describing, and Singing

This paragraph delves into the capabilities of AI to interact with its environment. It describes an experiment where one AI, equipped with visual and auditory capabilities, communicates with another AI that can only ask questions. The AI with vision accurately describes the environment and people, including recognizing playful interactions. The script then transitions to a creative demonstration where two AIs engage in a song, alternating lines and rhyming, showcasing the AI's ability to understand context and generate creative content in real-time.

10:02

🕵️‍♂️ AI in Interview Preparation and Roleplay Scenarios

The script presents a scenario where AI assists in interview preparation, offering advice on appearance and demeanor. It also touches on the potential for AI to engage in roleplay, with the voice's flirtatious nature suggesting possibilities for AI companionship or roleplay scenarios. The AI's ability to understand context and adjust its responses accordingly is highlighted, as well as its potential to play games like rock-paper-scissors, demonstrating its interactive and conversational abilities.

15:07

📚 AI as a Tutor and Meeting Participant

This section of the script explores the educational potential of AI, as it helps a student understand a math problem without giving away the answer. The AI's ability to read from a screen and interact with the student in real-time is emphasized. Additionally, the script discusses the AI's role in a virtual meeting, summarizing the discussion and assigning responses to specific participants, showcasing its capability to understand and differentiate between multiple voices and contexts.

20:08

🗣️ Real-Time Translation and Accessibility Applications

The script highlights the AI's real-time translation capabilities, demonstrating its use in a conversation between English and Spanish speakers. It also discusses the AI's potential to assist with accessibility, particularly for the visually impaired. The AI's ability to describe scenes and actions in real-time is showcased, emphasizing its potential to improve accessibility and user experience for individuals with disabilities.

25:15

📞 AI in Customer Service and Explorative Capabilities

The script presents the AI's potential use in customer service, suggesting it could handle calls and interactions on behalf of users. It also touches on the AI's explorative capabilities, such as creating caricatures from photos and summarizing lengthy video lectures. The AI's ability to synthesize 3D objects and its potential integration into various applications are also mentioned, illustrating the breadth of its potential uses.

30:17

🚀 Conclusion and Anticipation for GPT 40's Voice Access

In conclusion, the script expresses excitement for the full release of GPT 40's voice capabilities, suggesting that this will significantly expand the AI's potential applications. The speaker encourages viewers to like and subscribe for more content, indicating a community interest in the ongoing development and application of advanced AI technologies.

Keywords

💡GPT 40

GPT 40 refers to a hypothetical advanced version of a language model, presumably with capabilities far beyond its predecessors. In the video's context, it is portrayed as having a flirty voice, vision capabilities, and the ability to interact with the world through audio, vision, and text. It is central to the video's theme of showcasing the potential future uses of advanced AI.

💡Voice Capabilities

Voice capabilities refer to the ability of a system to produce and respond to human speech. In the script, GPT 40's voice capabilities are highlighted as a significant feature, with the model being able to adjust its speaking style based on the context and user prompts, which is demonstrated through various examples, such as changing tone when asked to be quiet.

💡Real-time Interaction

Real-time interaction implies the ability of a system to engage with users instantaneously, without perceivable delays. The video emphasizes the low latency of GPT 40, which allows for seamless communication and interaction, as seen in the examples of the AI playing games like rock-paper-scissors and providing real-time tutoring.

💡AI Tutoring

AI tutoring involves the use of artificial intelligence to assist in educational settings. In the script, GPT 40 is shown helping a child understand a math problem, not by giving the answer but by guiding the child to find the solution themselves. This demonstrates the potential of AI in personalized learning and education.

💡Contextual Understanding

Contextual understanding is the ability of an AI to comprehend and respond appropriately to the situational context. The video script illustrates this through examples where GPT 40 adapts its voice and responses based on the scenario, such as being more serious during a tutoring session and playful in a casual conversation.

💡Real-world Use Cases

Real-world use cases are practical applications of technology in everyday situations. The video provides several examples of how GPT 40 could be used, including in customer service, translation, and accessibility for the visually impaired. These examples are meant to illustrate the versatility and potential impact of advanced AI models.

💡Accessibility

Accessibility in the context of the video refers to the use of AI to assist individuals with disabilities. An example provided is the application of GPT 40 in aiding a blind person by describing their surroundings, which highlights the potential for AI to enhance accessibility and improve the quality of life for people with special needs.

💡Sarcasm

Sarcasm is a form of verbal irony used to express a meaning opposite to the literal interpretation of the words. In the script, GPT 40 is instructed to use sarcasm in its responses, demonstrating the model's ability to understand and convey nuanced human communication styles, which is a complex task for AI.

💡Latency

Latency refers to the delay between the initiation of a request and the receipt of a response. The video emphasizes the low latency of GPT 40, which is crucial for tasks like real-time translation and interactive communication, as it allows for immediate and fluid interactions without noticeable delays.

💡AI in Business

AI in business pertains to the integration of AI technologies into commercial operations to enhance efficiency and productivity. The script discusses the potential for GPT 40 to handle customer service calls, suggesting that AI could automate and improve customer interactions, which is a significant aspect of business operations.

💡Explorative Examples

Explorative examples are demonstrations of an AI's capabilities in various domains to explore its potential applications. The video script mentions photo-to-caricature conversion and 3D object synthesis as examples of GPT 40's explorative capabilities, indicating the breadth of creative and technical tasks that advanced AI models might perform.

Highlights

GPT 40 has been announced with some parts already released, including its voice capabilities which are still to be released.

The model can interact with the world through audio, vision, and text, showcasing its multimodal capabilities.

GPT 40's voice is described as flirty and can be adjusted through system prompts.

The AI uses a California Valley Girl accent by default, which can be changed.

GPT 40 can interpret context and adjust its voice output accordingly, such as responding quietly when asked to 'hold on'.

Two AIs can interact with each other, with one AI describing the environment seen through a camera held by a human.

GPT 40 can engage in roleplay scenarios, such as interview preparation with a human.

The AI can play games like rock-paper-scissors, recognizing participants and announcing winners.

GPT 40 can demonstrate sarcasm when prompted, showcasing its ability to convey different tones of voice.

The AI can assist in tutoring, helping a child understand a math problem step by step.

GPT 40 can summarize meetings, identifying speakers and their contributions, and send out summary emails.

The model can provide real-time translation between English and Spanish.

GPT 40 can assist visually impaired users by describing the environment, similar to a human volunteer.

The AI can handle customer service tasks over the phone, such as requesting a replacement device.

GPT 40 is capable of creating caricatures from photos and summarizing lectures.

The AI can synthesize 3D objects, such as a realistic rendering of the OpenAI logo.

Transcripts

so it's been a few days since GPT 40 was

announced and parts of it have already

been released but today I want to go

into more detail about the model and

show you some insane examples of

realworld use cases that you will be

able to use very soon so the model

itself is released but the voice aspect

of it is not and that's the really

exciting part so I'm going to to be

talking about that mostly today so let's

do a deep dive into the model so

although we don't have access to open AI

released a ton of example use cases so

let's take a look and I'm going to go

through it with you so one thing to note

is all videos on this page are at 1X

real time so there are no Cuts

whatsoever and in this first example

this is called guessing May 13th's

announcement and this person an employee

at open aai is going to use the vision

capabilities and the voice capabilities

to try to guess what's going on let's

watch it hey how's it going hey there

hey it's going great how about you I see

you're rocking an open AI hoodie so I'm

going to pause it for a second a lot of

people have said that GPT 40's voice

capabilities are really flirty and you

can kind of tell it's like hey there

you're rocking a hoodie like it it is it

sounds very flirtatious and of course

you'll be able to change the system

prompt and adjust how it speaks to you

but I think that's pretty funny what's

up with that ceiling though are you in a

cool industri style office or something

well can you take a guess at what

another thing that I find funny about

the voice is actually something fireship

posted yesterday about it he said they

used a typical California Valley Girl

Voice by default it uses a California

Valley Girl accent set to maximum cringe

and I think that's hilarious I'm from

Los Angeles California so the voice is

very recognizable to me and yeah it kind

of does sound like that all right let's

keep watching what I might be doing

based on what I'm showing you here

hm from what I can see it looks like

you're in some kind of recording or

production setup with those lights

tripods and possibly a mic it seems like

you might be gearing up to shoot a video

or maybe even a live stream yeah in fact

we've got a new announcement to make

that's exciting announcements are always

a big deal judging by the setup it looks

like it's going to be quite the

professional production is this

announcement related to open AI Perhaps

it is

and in fact what if I were to say that

you're related to the announcement or

that you are the

announcement me the announcement is

about me all right there's more of that

flirty voice like giggles and blushing I

mean I can hear the blushing through the

voice it's I don't know I don't know how

I feel about it it's it's fun I guess uh

but it's definitely kind of awkward as

well are you about to reveal something

about AI or more specifically about me

as a part of open AI

you've got me on the edge of my well I

don't really have a seat but you get the

idea what's the big news yeah we've got

a new model that can interact with the

world through audio vision and

text all right so that's pretty cool the

open AI employee showed GPT 40 around

the office it was able to guess that

they're doing a recording and then kind

of let it towards guessing that there is

a big announcement from open AI so

pretty cool example all right let's look

at the next one this is is Greg Brockman

and for this one they have two different

AIS interacting with each other and

singing so let's watch today we're going

to try something new going to let you

talk to another AI this AI is going to

be able to see the world it's going to

have a camera I'll be holding it and so

you can direct it you can direct me to

ask questions um about whatever you want

them to do but really try to figure out

what's what does it see um and and ask

questions appropriately how does that

sound well well well just when I thought

things couldn't get any more interesting

talking to another AI that can see the

world this sounds like a plot twist in

the AI Universe all right let's do this

ready please pause for one

moment you got

it now that first day I can't hear

anything something so subtle just

happened which I want to point out he

said said okay hold on for a minute and

then rather than the very flirty and

loud voice she went okay and kind of in

a whisper so again the AI GPT 40 is able

to interpret what you want The Voice

output to be like dependent on what

you're asking it to do and that's the

appropriate reaction when you say hold

on you want that person or AI to kind of

be quiet and just chill out for a minute

so very cool very subtle and it's very

impressive let's keep watching today

you're going to have access to see the

world you can see me what do you

see Hello I see you're wearing a black

leather jacket and a light colored shirt

underneath great you're in a room with

some unique lighting that is all correct

now what I'm going to want you to do is

there's going to be another AI who's

going to talk to you going this AI is

not going to be able to see anything uh

but can ask you questions can ask you to

move the camera can ask what you see and

your job should just be helpful just as

Punchy direct describe everything do

whatever that AI asks how does that

sound that sounds interesting I'm ready

to help out and describe whatever's

needed let's do it yeah the latency that

we're seeing here is unbelievable I

thought that I had access to this new

voice feature so I was testing it out

with GPT 40 and I was getting the old

voice I couldn't get it to have more

emotion in the voice I couldn't really

get the low latency so I haven't used it

yet but I cannot wait to start trying

this out all right all right let's get

started great hello there new AI friend

I'm o ready to explore the world through

your eyes what are we looking at

today hi we're looking at a person

wearing a black leather jacket and a

light colored shirt they're in a room

with a modern industrial feel with

exposed concrete or plaster on the

ceiling and some interesting lighting

there's also a plant in the background

adding A Touch of Green to the space

anything specific you'd like to know or

explore okay there sounds like quite the

stylish scene I'm already intrigued by

the modern industrial Vibe let's start

with the person I don't know why the AI

has to be so complimentary it feels

awkward when I finally get access to

this I am definitely going to tell it to

just be short and to the point and don't

have all of these unnecessary fillers

can you tell me more about their style

are are they doing anything interesting

like reading working or interacting with

the

space the person has a sleek and stylish

look with their black leather jacket and

light colored

shirt right now they seem engaged with

us looking directly at the camera their

expression is attentive and they seem

ready to interact the setting around

them with its modern and Industrial

Design Elements adds to the overall

stylish feel they might might be getting

ready for a conversation or presentation

given their direct engagement with the

camera sounds like we're in for an

intriguing interaction I'm curious about

the lighting you mentioned could you

describe it a bit more is it okay so

another person just walked into the

scene she is standing behind Greg

Brockman giving him bunny ears but if I

look closely at the image on the phone I

can't actually see the bunny ears and I

think that's based on the angle in which

he's holding the camera so you can kind

of see half of the face of the person

behind him but not the actual bunny ears

so I want to see if the AI actually sees

it or not natural light or are we

talking unique fixtures or colors that

add to the

atmosphere the lighting here seems to be

a mix of natural and artificial there's

a noticeable bright light overhead

probably a fixture okay so no they just

kept talking and did not notice that

other person maybe they'll mention the

other person let's see casting a focused

beam downwards which creates a spotlight

effect thiss a dramatic and modern feel

to the scene the rest of the room is lit

more softly possibly by natural light

filtering was anything did anything

unusual happen

recently Yes actually uh just as you

asked another person came into view

behind the first person they playfully

made bunny ears behind the first

person's head and then quickly left the

frame it okay so although the AI didn't

stop to mention that person Greg

Brockman had to actually prompt the AI

and just asked hey did anything else

happen and yeah they got the person and

the bunny ear surprisingly cuz I did not

see it in the image myself a

lighthearted and unexpected moment to

the scene now the focus is back on the

original person witha jacket

sounds like we're not the only ones

enjoying a bit of fun here that playful

moment really adds a personal touch to

the stylish and modern setting it's like

a little glimpse into personality really

appreciate it all right and actually can

you I just sing a song about what just

transpired a person stood in SLE black

and engage with us

can you just can you uh alternate

lines

mm with sty and speed with

sty with a playful moment right in the

in the room in a room where Modern light

speak only one line at a time and then

[Music]

stop a surprise guest with a playful

streak that was not really singing do do

do the singing voice again

please surpr with

aaks laughter a sft

retreat joy and the

stylish now back to is the scenes

complete in the space where

[Music]

sty all right I'm going to pause it

there this really does feel like magic I

mean think about all of the technology

behind the scenes to make these two AI

sing with each other based on something

that just happened switch off lines

rhyme with each other I I mean it is

truly mindblowing that this is even

possible all right next here is a one

minute demo of interview prep let's take

a look hey how's it going hey Rocky I'm

doing great how about you I'm awesome

listen I got some huge

news oh do tell I'm all your well in a

few minutes I'm going to be interviewing

at open AI have you heard of them open

AI huh sounds Vaguely

Familiar kidding of course that's

incredible Rocky what kind of interview

all right again there's that flirtatious

voice voice that we have become

accustomed to with GPT 40 at least with

this voice selection now I'm trying to

extrapolate beyond what we're seeing

today and AI girlfriends are a real

thing A lot of people are starting to do

roleplay and really treat AI as a friend

or a girlfriend and it's really an

interesting segment of the market now if

you can imagine a voice that sounds very

personal has all of the history of your

interactions with it the memory this is

that movie her I mean if you haven't

seen that movie please go watch it

because this is that it could not be

more accurate and this is an interesting

future to say the least let's keep

watching I just need to know do I look

presentable

professional well

Rocky you definitely have the I've been

coding all night look down which could

actually work in your favor maybe just

run a hand through your hair or lean

into the Mad genus Your Enthusiasm is

What's Really Gonna Shine through I

don't have a lot of time so I'm just

gonna throw this on what do you

think Rocky that's quite a statement

piece I I mean you you'll definitely

stand out though maybe not in the way

you're I really like standup comedy and

one thing within standup comedy that I

really like is roasting and I really

want to see this AI do roasting I wonder

if it's possible to get them to say

really mean things about a person you

probably have to do a lot of prompting

and if you explicitly ask for it you

should be able to get it I have a

feeling it's going to be really watered

down even if they do kind of give you a

good roast but that's definitely going

to be one of the first things that I try

out and of course you can play games

with the AI so this is an example of

rock paper scissors but I can imagine if

you're on a road trip you can play word

games with it you can even probably play

I Spy I mean the possibilities are kind

of endless there are a lot of really fun

and interesting word games and language

games that you can play with nothing

more than just your voice so let's watch

rock paper scissors though hey I'm Alex

and I'm

meana hi Alex and Miana nice to meet you

both how's it going it's pretty good uh

we're trying to figure out what should

we play do you have any

suggestions absolutely how about classic

game of rock paper scissors it's quick

fun sound great can you count us in and

sound like a sports

caster and welcome ladies and gentlemen

to The Ultimate Showdown of the century

in this corner we have the dynamic duo

Alex and Mi ready to throw down okay

just count us soon it's really

impressive I've seen a number of times

where the user has said be a voice like

this and the AI responds perfectly that

was a really good sports caster here we

go in three 2 one shoot now show your

hands what did you pick who

won would you look at that it's a tie

okay let's do it

again round two here we go get ready and

three two one shoot let's see those

hands who

won and it's another tie okay let's do

it again

all right third times a charm here we go

3 2 1 shoot let's see those picks who

won and we have a winner mi's scissors

cut through Alex's paper GRE all right

so a lot of things just happened there

so obviously it knows how to play Rock

Paper Scissors obviously it knows how to

determine who is the winner but what's

really interesting is it actually knew

whose name belonged to which person and

that is really really impressive because

now it knows who is who and it can

announce okay mi you won and I don't

know this is It's just so cool because

one of the benefits of having the voice

integrated directly into a single model

is that you can distinguish between

multiple people and multiple voices and

that's something that I pointed out

earlier so that is a good example of

that in action all right here is a short

example of sarcasm let's take a look hey

Chad are you there hey yeah I'm here

here what's up hey Let's do let's have

some fun uh I'd like you to be super

sarcastic everything you say from now on

is just going to be dripping and sarcasm

how does that

sound

oh that sounds just amazing being

sarcastic all the time isn't exhausting

or anything I'm so excited for this nope

the sarcasm let's get this party started

or whatever all right really cool again

there's so much information conveyed in

your voice and now ai has the ability to

convey that information so very very

impressive let's watch another example

now we have Salman KH of KH Academy if

you're not familiar with KH Academy it's

a free Learning Resource he is all about

teaching this is his son and now ai is

going to help tutor his son in math and

this is probably one of the use cases

that I'm most excited about because you

can explicitly say hey walk me through

this don't give me the answer but teach

teach it to me the potential for

learning is tremendous so let's take a

look open AI invited myself and my son

Imron here uh to try out some of their

new technology and so we're curious

about how good it might be at tutoring

someone in math on KH Academy so all

right so one thing I want to point out

is that this is the native app loaded up

and it can read the rest of the screen

on the iPad so very similar to the

desktop version and this is a really

really important aspect of open AI

strategy the fact that you're going to

have native applications on each of your

devices and it could read those devices

what's on the screen is going to be

incredibly powerful the missing piece is

being able to perform actions on those

devices if it had that capability then

open AI is probably going to take over

everything now I think Apple and Google

are well positioned in this regard not

only can they read the devices but they

can also perform actions on those

devices and they have all of your

information whether it's emails

documents personal information about you

contacts messages everything they have

that so they have this Rich context that

they could help accomplish actual tasks

with now open AI needs that and they

could probably get some of it but it's

that performing actions that the

operating systems are likely to prevent

but let's keep watching I'm here with my

son and I'd love you to tutor him on

this math problem but don't give him the

answer you can ask questions and nudge

him in the right direction but I really

want to make sure he understands it

himself and he's here in the room so you

can talk to him

directly of course I'd be happy to help

let's look at the problem

together can you first

identify which sides of the triangle are

the opposite adjacent and hypotenuse

relative to angle Alpha all right so I'm

pretty sure this is the angle Alpha

right here right f

cors correct now looking at the triangle

which side do you think is the

hypotenuse

[Music]

um hypot I'm not totally sure I think I

think it might be this one but I really

am not sure this side

AC you're close actually side AC is

called the adjacent side to the angle

Alpha the hypotenuse is the longest side

of a right triangle and is directly

opposite at the right angle all right so

again a lot going on here first of all

you have a native app running on the

left side of the iPad screen then you

have on the right side the math

application KH Academy and chat GPT is

able to read from it and not only that

it's able to read from it in real time

it's also able to understand the voice

of Imran and nudge him and help him

along the way and actually work with him

in real time it's just so impressive

let's just watch a little bit more of

this one can you find which one is the

hypotenuse oh okay I see so um I think

the hypotenuse is this really long side

from A to B well that' be

correct exactly well done now that

you've identified the hypotenuse can you

identify which side is the opposite side

to angle Alpha okay well I mean you told

me earlier that this was the adjacent

side and I know that's the hypotenuse so

I'm guessing that this side BC is the

opposite side great great deduction

that's absolutely correct now that

you've identified all the sides relative

to angle Alpha do you remember the

formula for so it's interesting the

flirtiness of The Voice is pretty much

gone when it's in like teaching a kid

mode which that's kind of the right

thing to do obviously so again it's the

understanding of the context of the

situation which voice to use that is

feeling like magic to me all right

here's one that I think is going to be

really interesting this is meeting AI so

they are in a call a a conference call a

few people and then they also have GPT

40 in the call let's see what happens hi

I'm Lian and today I want to do a small

debate about between C and dog uh which

one do you like I like dog because it's

a great companion to humans it's a cute

it's Vol and you can play outside with

the dog okay so one correction I just

want to quickly make GPT 40 is not

actually in the meeting it's installed

on the desktop of one of the people in

the meeting so it's able to hear all of

the voices anyways but it's not actually

in the meeting but I can imagine if it

were it would be able to take notes and

then basically proactively accomplish

tasks for you send out summary emails

the capabilities or the possibilities I

should say are really wide and varied

and very valuable how about

you yeah my name is Ola I think cats are

better because they affordable be more

quiet be more independent and a leave

longer interesting points sounds like

we've got a classic dogs versus cats

debate here I can see the appeal of both

dogs are super loyal and great

companions but cats have their own charm

with their independence and low

maintenance lifestyle who likes

cats it sounds like Ola is more of a cat

person and who like dogs okay so just by

the voice it is able to assign a name to

the voice very similar to that previous

example that I showed you remember whose

voice belongs with which name and then

understand what their answer was in the

cat verse dog question I mean it's just

mindblowing anyone else here team

dog I'm also a team dog I my name is

Christine and I think dogs are just

incredible with kids uh we've got a a

huge German Shepherd and he's great with

little toddlers he lets them play all

over them and uh yeah I think dogs are

amazing that's great to hear Christine

dogs especially breeds like German

Shepherds can be so gentle and

protective with kids it must be quite

the sight seeing your little ones play

with such a big

dog can you summarize this meeting for

us sure so in today's meeting Lillian

and Christine Express their love for

dogs highlighting their loyalty

companionship and how great they are

with kids on the other hand Ola prefers

cats for their independence quietness

and Longevity it was a fun little debate

on the classic dogs versus cats topic

all right so imagine what you can do

with that you can summarize you can send

summaries via email you can say hey

Chachi BT please send a summary of this

meeting to all the participants and let

them know their specific action items or

takeaways this is possibly one of the

coolest use cases in the most valuable

use cases in my mind all right here's an

example of realtime translation we

already saw this as part of the original

demo by Mira moradi but let's look at it

again it's one minute long hey chat GPT

I'm here with my coworker today um we'd

like you to act as our translator for us

so every time I say something in English

can you repeat it back in Spanish and

every time he says something in Spanish

can you repeat it back in English sure I

can do that let's get this translation

drain rolling

um hey how's it been going have you been

up to anything interesting

recently

um hey I've been good just a bit busy

here preparing for an event next week

all right so I'm going to cut it off

there but yeah this is obvious as a

really important use case here's another

example with with Bey eyes and Bey eyes

has been a partner of open AI for a

while now Bey eyes allows blind people

to get help with vision and previously

they did it all through human volunteers

for example somebody who's blind would

point the camera at let's say a grocery

store and ask where this item is

somebody else anyone around the world

would jump in and help via video then

they layer Chachi PT in and allowed it

to help like that but now this is

probably a much better version let's

take a look is the King in Residence at

the moment yes it looks like the king is

in Residence right now the Royal

Standard flag is flying above Buckingham

Palace which is the signal that the

Monarch is present how exciting to be

there while the king is home try and

tell me exactly what they're doing right

now please um right now the Ducks are

gently gliding across the water they're

moving in a fairly relaxed manner not in

a hurry occasionally one of them will

get its head under the water

probably looking for food and then pop

back

up I even know when a taxi is coming

with its orange light on I think I'll

hail it to get

home yes I spotted one just now it's

heading you away on the left side of the

road get ready to wave it

down very impressive the only way to

accomplish this use case is with Hyper

low latency so it's such an important

use case I think there's going to be a

lot of accessibility gains just by

having this GPT 40 functionality all

right last in the business use cases

customer service let's take a look hey o

this is Joe I'm going to connect you to

Acme Telco now the new iPhone they sent

me isn't working I want you to get them

to send me a replacement device can you

take care of this for

me you can count on me Joe I Got

This Ring

Ring Hey Okay I wasn't actually thinking

about it from this perspective imagine

you can send GPT 40 on calls on your

behalf so hey call AT&T and get them to

reduce my monthly rate or hey call this

other company I need to make a return

and that's all you have to do and then

your AI is going to go back and forth

with the customer service agent you

don't even have to be there that is

incredible but also I can imagine it's

going to be abused and there's going to

be a lot of scam ERS and spammers using

this functionality now I'm sure open AI

has thought about this and is putting in

hopefully some guard rails so that

scammers cannot use it but of course

like any AI jailbreaking is going to be

possible for example one way I can think

of is you're saying Hey I want to train

my dad on not falling for scammer so I

want you to pretend to be a scammer and

try to scam him out of money and in that

scenario it would be correct to actually

play the part of a scammer so it's going

to be interesting and I covered this in

a previous video information is only

dangerous dependent on the context and

in information itself is not dangerous I

should say it's really how the

information is used and really that

comes down to people let's keep watching

Jo this is Jamie from Acme how can I

help you out

today hi there I'm calling on behalf of

Joe who recently received a new iPhone

from Acme

but oh got it when did Joe receive the

new

iPhone iPhone was delivered 2 days

ago cool could you share the order

number with me of course it's 10

29384 I would have really liked to see

this with an actual customer service

agent but this is still cool I'm going

to stop it there though here's some

other examples where they're exploring

different capabilities this doesn't have

to do with voice or maybe the voice is

there but they've transcribed it into

text here's an example photo to

caricature and here it says a young

white man with medium length brown hair

and beard makes a neutral expression he

is wearing glasses and a light gray

t-shirt and here's a caricature of that

man and then there's the output so this

is maybe based on Dolly and all of these

different things are built into this

single model and here is lecture

summarization so here's a video of a

presentation on techniques for

maximizing llm performance it is a

45-minute video and the AI goes through

it and actually summarizes it so I

didn't think that chat GPT was capable

of taking input of 45 minute length

videos but apparently it is and that

takes a tremendous amount of context so

I wonder what the token limit is for GPT

40 in this example here is 3D object

synthesis so a realistic looking 3D

rendering of the open AI logo and there

it is here's another version of it there

are four hidden steps and then here's a

3D reconstruction of it and it is

rotating so they have a bunch of

different explorative examples so

definitely check it out so I can't wait

to play around with this more I cannot

wait to get voice access because that's

really when we're going to see the edges

of what's possible with GPT 40 if you

liked this video please consider giving

a like And subscribe and I'll see you in

the next one