AI News: This Was an INSANE Week in AI!
Summary
TLDRThis video discusses a series of major AI announcements from OpenAI and Google. OpenAI's spring event introduced GPT-40, a multimodal model capable of understanding text, audio, video, and images, which is now available to all ChatGPT users for free. They also showcased its enhanced conversational abilities and new desktop app. Google IO revealed a multitude of AI innovations, including the Gemini 1.5 models and Project Astra's interactive capabilities. The video also covers other industry updates, such as departures from OpenAI and new roles at Anthropic, and previews upcoming AI-related events.
Takeaways
- 🚀 This week has been full of AI news, primarily from OpenAI and Google.
- 📢 OpenAI announced GPT-4O at their spring event, a multimodal model that handles audio, video, and images.
- 🎉 The new GPT-4O model is available for free to all ChatGPT users, with advanced features like the GPT store, vision models, and internet browsing.
- ⚡ GPT-4O API is twice as fast, 50% cheaper, and has a five times higher rate limit compared to the previous model.
- 💬 The conversational capabilities of GPT-4O are impressive, likened to the movie 'Her' with a voice similar to Scarlett Johansson's.
- 📱 A new desktop app for GPT-4O will be available on Mac first, with the ability to see and interact with the user's screen.
- 🖼️ GPT-4O includes an image generation feature, capable of creating realistic images and maintaining character consistency across different scenes.
- 📉 Google IO introduced many AI announcements, including the Gemini 1.5 models with a 1 million token context window and Project Astra for enhanced mobile phone interaction.
- 🎥 Google's Imagine 3 and Veo video generation models aim to compete with other advanced AI image and video generation tools.
- 📅 Upcoming events from Microsoft, Cisco, Qualcomm, and Apple are expected to bring more AI-related announcements and innovations.
Q & A
What were the main events that dominated the AI news this week?
-The main events were OpenAI's spring update and Google's annual IO event.
What significant model did OpenAI announce during their spring update?
-OpenAI announced GPT-40, a multimodal model that can handle audio, video, and images.
What is unique about GPT-40 compared to previous models?
-GPT-40 is faster, multimodal, understands audio, text, images, and video, and is available to free ChatGPT users.
What are some of the new capabilities of GPT-40?
-GPT-40 can generate images, recognize facial expressions, help solve math problems, and interpret data from screens.
How is GPT-40 API beneficial for developers?
-The GPT-40 API is two times faster, 50% cheaper, and has a five times higher rate limit compared to previous models.
What major change happened in OpenAI's leadership recently?
-Ilia Sutskever, one of the original founders of OpenAI, decided to step away from the company.
What new AI feature did Google demonstrate at their IO event that involves a mobile phone?
-Google demonstrated Project Astra, which allows a phone to see what it's looking at and answer questions about it.
What is the context window capability of Google's new Gemini models?
-The Gemini models have a 1 million token context window, with plans to increase it to 2 million tokens.
What was the highlight of Google's announcements in terms of video generation?
-Google introduced Veo, a new video generation model capable of creating 1080p videos over one minute long.
What was Anthropic's notable recent hire and their role?
-Anthropic hired Instagram's co-founder Mike Krieger as its Chief Product Officer to improve user interfaces and experiences.
Outlines
📰 Overview of AI News Week
This week was packed with AI news primarily from Google and OpenAI. OpenAI had their spring event on Monday, followed by Google’s IO event on Tuesday. Highlights include the announcement of GPT-4.0, a multimodal model handling audio, video, and images. OpenAI’s updates make advanced features available to free users. Key features include a faster GPT-4.0 API and conversational capabilities resembling the movie 'Her'. OpenAI also introduced a new desktop app with screen recognition and the ability to interpret and debug code.
📸 GPT-4.0 Capabilities Beyond the Keynote
OpenAI's blog revealed more about GPT-4.0’s capabilities not shown in the keynote, such as image generation, consistent character rendering in different contexts, and creating 3D object syntheses. Examples include generating legible text on a whiteboard, maintaining character consistency across different scenes, and creating 3D renderings from multiple images. These capabilities showcase GPT-4.0's potential in various creative and practical applications.
🎙️ Sam Altman on Using GPT-4.0
Sam Altman discussed practical use cases for GPT-4.0 in an interview on the Logan Bartlett Show, emphasizing how it enhances productivity by providing instant responses without switching tasks. He also addressed the departure of OpenAI co-founder Ilia Sutskever, speculating it was due to disagreements over the company's direction. The departure of other key members of the superalignment team suggests internal shifts at OpenAI.
🔍 Google IO Announcements
At Google IO, the company introduced the Gemini 1.5 Flash and Pro models, capable of handling up to 2 million tokens. They also showcased Project Astra, which can answer questions about visual inputs and remember previous observations. Another major reveal was the potential integration of these capabilities into AR glasses, hinting at the next generation of Google Glass. Google’s AI-driven features are expanding across its suite of tools, enhancing productivity and user experience.
🎥 Google’s Veo Video Generation
Google introduced Veo, a new video generation model producing high-quality 1080p videos, with features surpassing other models like Sora. Notably, Veo will be available to the public soon, with a waiting list already open. Google’s integration of AI into search and email enhances user convenience by summarizing information and organizing data efficiently.
🔊 Google’s Enhanced AI Tools
Google’s updates extend across its services, including AI-generated summaries in Gmail and Google Meet, and the ability to organize and analyze data in Google Drive. The 'Ask Photos' feature provides contextual information about personal photos. Google also showcased a scam-detection feature in phone calls, alerting users of potential fraud in real-time.
🎙️ Interactive AI News Podcast
Anthropic hired Instagram’s co-founder to enhance user interfaces for its tools. They released a new prompt generator for better prompt engineering. Meanwhile, the company Hume introduced 'Chatter,' an interactive podcast where users can steer the conversation by asking questions, enhancing engagement with AI-driven content.
👂 Meta’s AI Earphones and Future Events
Meta is exploring AI-assisted earphones with cameras to provide contextual awareness. The upcoming Microsoft Build event promises more AI announcements, likely integrating GPT-4.0 into various products. June will also feature AI and cybersecurity events by Cisco and Qualcomm, and Apple’s Worldwide Developer Conference, all focusing on AI advancements.
Mindmap
Keywords
💡AI News
💡GPT-4
💡OpenAI
💡Google IO
💡API
💡AI Assistant
💡Image Generation
💡Multimodal
💡AI Test Kitchen
💡Anthropic
💡Interactive Podcast
Highlights
OpenAI announced GPT-40, a multimodal model that can handle audio, video, and images, and is available for free ChatGPT users.
GPT-40 is faster, cheaper, and has a higher rate limit for developers using the API.
The new model's conversational capabilities are reminiscent of the movie 'Her,' with voice interactions similar to Scarlett Johansson's voice.
GPT-40 can help users solve math problems step by step using the camera.
OpenAI introduced a new desktop app for Mac, with future plans for PC, that can see what's on the screen and assist with coding and debugging.
GPT-40 can recognize facial expressions and guess the user's emotions.
OpenAI's blog post reveals additional capabilities of GPT-40, including image generation, character consistency, and 3D object synthesis.
Ilia Sutskever, a co-founder of OpenAI, announced his departure, citing a desire to pursue a personally meaningful project.
Google introduced Gemini 1.5 Flash and Pro models, with a million-token context window, and a potential future increase to two million tokens.
Project Astra from Google can analyze what a mobile phone camera sees and answer questions about it, even remembering past observations.
Google's new Notebook LM feature can organize data and create interactive podcasts from user inputs.
Google's Veo video generation model competes with Sora, creating realistic videos up to one minute long.
Google Search now offers multi-step reasoning capabilities and can answer complex queries comprehensively.
Anthropic hired Instagram's co-founder as its head of product and introduced a new prompt generator for creating effective, precise prompts.
Microsoft's upcoming Build event is expected to integrate GPT-40 into its products, enhancing their capabilities.
Transcripts
so this week kicked off a series of
insane weeks where we're about to just
get bombarded with AI news this week
there was a ton now if we're looking
closely here most of it came out of
either Google or open AI seeing as how
open AI had their spring event on Monday
and Google had their annual IO event on
Tuesday it's pretty much been a week
that I can't really describe as anything
else but a giant announcement Fest the
week kicked off off on Monday with open
AI spring update I actually did a full
breakdown video of that spring update on
the channel you can find it it's called
chat gpts amazing new models feel human
and it's free however I do know a lot of
people watch just this one end of the
week video on this channel to get
totally caught up with all the AI news
so I will give you a quick recap plus
I'll share a few other things that they
didn't talk about in that keynote that
are pretty cool the event kicked off
with the announcement of GPT 40 the O
stands for Omni because it's a
multimodal model that can handle audio
and video and images and kind of
everything you can throw at it last week
when we were all talking about this
mysterious gpt2 chatbot that was on the
lm.org website and we were all
speculating that it was a new open AI
model well you're right this is what GPT
40 was now this new GPT 40 model is much
faster it's multimodal it understands
audio text images video and just
generally seems a lot better of a model
than what we were getting out of the
previous gp4 model but one of the big
announcements that they made at this
event was that it's going to be made
available for all the free chat GPT
users so now you don't even need to have
a paid plan of chat GPT to get access to
the GPT store the vision models the
browsing the internet models the memory
the Advanced data analytics AKA code
interpreter all of this is now available
inside of the free chat GPT model now if
you are a chat GPT plus member you're
still going to get features a little bit
earlier than the free members and you
get I think five times more outputs from
chat GPT so you can use it a lot more
but free users now get access to all of
the cool features that paid members have
gotten for those developers out there
the GPT 40 API is actually two times
faster than the last Model and it's 50%
cheaper and has a five times higher rate
limit all those developers out there
building with open AI GPT for API you
just got a lot better model for less
expensive but it was the conversational
aspect of this new model that I think
really had people talking it's very
reminiscent of that movie her and even
the voice sounds a little bit like
Scarlet Johansson's voice from her
here's a few examples of what they
showed off hey Cat I'm Mark how are you
oh Mark I'm doing great thanks for
asking how about you hey so I'm on stage
right now I'm doing a live demo and
frankly I'm feeling a little bit nervous
can you help me calm my nerves a little
bit oh you're doing a live demo right
now that's awesome just take a deep
breath and remember you're the expert I
like that suggestion let me try a couple
deep breaths can you give me feedback on
my breaths okay here I go
whoa slow a bit there mark you're not a
vacuum cleaner they also showed off a
demo where they had a math problem and
then they used the camera to look at the
math problem and then it helped them
solve the math problem step by step it
didn't just give them the answer they
asked for help to solve it and so it
actually walked them through how to
solve it instead of just giving them the
answer and you can see it eventually
helped them get to the correct answer
they also showed off their new desktop
app which apparently is going to be
available on Mac first at PC sometimes
in the future but it works in the same
way that you use the mobile app where
you talk to the app however one huge
difference is that it can actually see
what's going on in your screen you can
see there's a little screen button here
and you can allow it access to see
what's going on in your screen to help
you with code or debug whatever you're
working on they showed it interpreting
some data on a graph here and then they
showed the capabilities of actually
changing how it speaks which is really
cool so cool that I think you actually
need to see the clip but they're
definitely trying to make it much more
emotive and I want you to tell him a
bedtime story about robots and love oh a
bedtime story about robots and love I
got you covered gather round Barrett
once upon a time in a world not too
different from ours there was a robot
named bite bite was a curious robot
always expl started this story but I
want a little bit more emotion in your
voice a little bit more drama got it
let's add some drama one upon a time in
a world not too different from ours
there was a robot named no chbt I really
want maximal emotion like maximal
expressiveness much more than you were
doing before understood let's amplify
the drama once upon a time in a world
not too different from ours there was a
robot named bite bite can you do this in
a robotic voice now initiating dramatic
robotic voice once upon time in a world
not too different from ours there was a
robot named bite they also show that it
can recognize your face and guess the
sort of emotions you're going through
based on your facial expressions it
looks like you're feeling pretty happy
and cheerful with a big smile and maybe
even a touch of excitement whatever's
going on it seems like you're in a great
mood care to share the source of those
Good Vibes and those were really the
main things that they shared during this
keynote presentation it's Worth to watch
it's only 26 minutes to see the whole
presentation and everything they showed
off however there were some things that
they didn't show in the presentation
that they talked about in their blog
post here they completely left out the
fact that GPT 40 is actually an image
generator as well here's an image that
they generated of somebody writing on a
whiteboard and the text is actually
legible and it's a lot of text too which
is not something we've seen really yet
but if you check out the blog post which
will be linked below the video called
hello GPT 40 they actually have a whole
bunch of demo videos of it doing
different stuff that they didn't show in
their main keynote you can watch some of
these videos to see things like the two
phones singing back and forth to each
other and helping somebody prep for an
interview and playing rock paper
scissors and seeing how good it is at
sarcasm and more math examples there's a
whole bunch of really cool demo videos
that show off similar stuff to what they
showed in the demo video there but if
you scroll down a little bit further
you've got this section called
explorations of capabilities and it's
got a drop down of all of these other
capabilities that it's got that they did
not show off in the keynote like poster
creation for the movie detective input
let's design another poster with two new
characters this is a picture of Alex
nickel and a casual picture of Gabriel
go they gave a description of what
they're looking for in the poster and it
and it generated this picture here it
botched their names but managed to get
both of their faces into the poster and
write the word detective on it they
asked it to do some cleanup and and it
got even closer look at that pretty cool
poster this was done with GPT 40 here's
another example that shows character
consistency they created Gary the robot
here and then prompted that same robot
into different scenes they uploaded the
original image that it generated and
then asked it to generate an image of
him playing frisbee and it created the
same character but playing frisbee or he
likes to program computers same
character at a computer riding a bike
cooking playing violin it all started
with generating a character and then
uploading the character and then it was
able to consistently use that same
character again this is really
impressive stuff like why wasn't this
stuff in the keyote here's a
commemorative coin that they made for
GPT 40 and then they actually had it
generate the sounds of coins clanging on
[Applause]
metal they have this photo to caricature
ability where they upload an image like
this and then it makes a caricature
version of that image or here's another
one where it turned that into this sort
of cartoony car image here's another
example and another example and another
example this looks like it's pretty
impressive and they were able to do this
with one single input image they have
text to font where you can describe a
font and it will actually generate
entire fonts for you 3D object synthesis
what how is this not in the main keynote
a realistic looking 3D rendering of the
open AI logo with open AI shown below it
created this it then created it looks
like five more images of it at different
angles and then animated it into this 3D
logo here here's another one of a sea
lion where it basically generated the
image at multiple angles and once it had
all of the angles it was able to create
a 3D reconstruction with all of the
images they have examples of brand
placement here where they uploaded the
open aai Logo and this coaster here and
then it actually transferred that logo
onto the coaster so after you watch that
open AI video where they show off the
speaking and it sounds like her and
they're talking back and forth and
having conversations and helping with
math definitely come check out the open
AI blog post watch some of these other
demos but then click through the
exploration of capabilities of some of
the other stuff they did with it because
this is so much more powerful than what
they actually LED on in their keynote on
Monday like they could have done what
Google did and just made announcement
after announcement after announcement
and showed off all this stuff one by one
but they kind of paired it down and just
showed you a few of the features which
is crazy because some of the stuff here
to me is actually more impressive than
what they showed in their demo on Monday
the day after their presentation Sam
Alman was interviewed on the Logan
Bartlett show and one of the questions
he was asked was what use cases have you
found for this new tech that you just
put out that has been really helpful for
you one surprising one is putting my
phone on the table while I'm like really
in the zone of working and then without
having to like change Windows or change
what I'm doing using it as like another
channel so I'm like working on something
I would normally like stop what I'm
doing switch to another tab Google
something click around or whatever but
while I'm like still doing it to just
ask and get like an instant response
without changing from what I was looking
at on my computer that's been a
surprisingly cool thing now GPT 40 and
all of the capabilities we just saw
weren't the only news to come out of
open AI this week in fact on Tuesday
during the big Google iio event we got
word that Ilia suser one of the original
Founders over at open AI who started the
company up with Sam Alman decid to step
away now if you remember back in
November of 2023 Sam Alman was fired and
then brought back in well Ilia was one
of the people that was on the board at
the time that made the decision to fire
Sam he then regretted his decision made
a public apology on X said it was a
mistake to get rid of Sam but then
nobody really heard from Ilia since
until this week when we got word that
Ilia was stepping away I can only
speculate at this point but my guess is
that he wasn't really a fan of the
direction that open AI is heading in now
Ilia himself is a researcher he's an
academic I believe he loves the science
and the technology behind it he's less
of a fan of the monetization
capitalization of the technology and I
think he sees the direction that's all
headed and deciding that that's not
really where he wants to go again just
speculation I have no conversations with
anybody at open AI so I don't know for
sure but but that's kind of the vibe
that's going around about it it seems
like Ilia left on good terms he said
after almost a decade I've made the
decision to leave open AI the company's
trajectory has been nothing short of
miraculous and I'm confident that open
AI will build AGI that is both safe and
beneficial under the leadership of Sam
Greg and Mira it was an honor and a
privilege to have worked together and I
will miss everyone dearly so long and
thanks for everything I'm excited for
what's to come next a project that is
very personally meaningful to me about
which I will share details in due time
so he is moving on to some other project
project he's just being kind of hush
hush about it right now Sam also took to
Twitter to say nice things about Ilia
Ilia and open AI are going to part ways
it's very sad to me Ilia is easily one
of the greatest minds of Our Generation
a Guiding Light of our field and a dear
friend open AI would not be what it is
without him and he has something
personally meaningful that he's going to
go work on they have a new Chief
scientist his name is Jacob Pachi but
again it seems like he left on good
terms who knows what's really going on
behind the scenes but after after Ilia
made the announcement that he was
stepping away we got word that more
people at open aai were also leaving
including key members of the super
alignment team the people that are there
to make sure that AI doesn't go Rogue
and try to kill us all essentially
according to this article Yan leak
Leopold Ashen brener and William
Saunders all quit now I'm sure I
butchered some of those names but that's
besides the point these are all people
that actually didn't sign the memo early
on when everybody was trying to get Sam
Alman back these people that left
according to this article on Gizmodo
here were some of the holdouts that
weren't signing notes to get Sam Alman
back into the company now these are the
same people that left right when Ilia
left now if I had to guess I think Ilia
made the decision to leave quite a while
ago and they sort of timed the decision
with the gp40 announcement and Google
build event and all that stuff that was
going on to sort of bury it a little bit
they didn't want this new of Ilia
leaving to be the main hype in the news
cycle right now so they sort of just
squeezed in the news among everything
else that was happening that's kind of
my feeling on what's going on here the
very next day after the open AI event
was Google IO Google's annual event
where they make a ton of announcements
and this year again the announcements
were all about AI it's pretty clear at
this point that open AI is strategically
planning their announcements to happen
right around the time Google does their
announcements because they always seem
to be trying to overshadow Google Now
While open AI sort of had one big
announcement they were really sort of
showing off GPT 40 and the voice
capability and how to use it like a
voice assistant Google took a different
approach and just bombarded us with AI
announcements in fact Google themselves
put out a blog post called 100 things we
announced at IO 2024 if you want to
listen to this article it's 21 minutes
long but again I also did my own
breakdown of this you can find it here
on YouTube called Google just took over
the AI World a full breakdown in that
video I walk through all of my thoughts
from Google IO and all of the various
announcements but for those of you who
this is the one video per week that you
tune into from me I'll give you the
quick highlights of what I thought was
the most interesting they introduced
Gemini 1.5 flash a new large language
model that's faster but maybe not always
the most intelligent so if you need it
to respond fast you use this model if
you need it to have the best possible
output you use the Gemini 1.5 pro model
both 1.5 Pro and 1.5 flash have a 1
million token context window with plans
to jump it to a 2 million token context
window and if you don't know what that
means one token is about 75% of one word
so 1 million tokens means between the
amount of text you input for the large
language model and the text you get back
in the response combined can be about
750,000 words when it jumps to a 2
million token context window you can now
input and output about 1.5 million words
now the biggest showstopper from the
event in my opinion was when they showed
off project Astra with project Astra
they were able to take a mobile phone
and it was able to see what the phone
was looking at and then you can ask
questions about what it was looking at
and even draw on your phone and get more
details here's an example of what that
looked like tell me when you see
something that makes
sound I see a speaker which makes
sound what is that part of the speaker
called that is the Tweeter it produces
high frequency sounds and so the phone
was actually watching everything that
the camera can see but it also
remembered everything that it saw during
this whole process so let me fast
forward this a little bit because there
was one really interesting part here
notice how she casually moves past the
glasses there well let me Zoom ahead a
little bit further and check this out do
you remember where you saw my
glasses yes I do your glasses were on
the desk near a red
apple and you can see it remembered
where those glasses are and here's a
little Easter egg that most people
probably didn't catch if you watch what
happens next this might be a sneak
preview of what Google's working on next
because she grabs the glasses here puts
them on and then you no longer actually
see her doing this with her phone it
appears from that point on she's doing
it with glasses they didn't specifically
talk about this in the keynote but if
you watch the video that's clearly what
happened so just watch the rest
here she puts her phone down puts on the
glasses now you see her wearing
glasses what can I add here to make this
system
faster adding a cach between the server
and database could improve speed and
notice how there's some text on the
screen there I'm assuming that's what
she's seeing in her glasses as she's
talking so it might have like a little
bit of a heads up display in the glasses
a little bit of an augmented reality
element to them but they haven't
actually confirmed this so I'm just
reading into this from seeing this in
the video let's take a peek down here in
the corner we can see her wearing these
glasses here this might be the next
iteration of Google Glass I mean Google
Glass was ahead of its time because now
we're seeing stuff like The Meta Ray
bands actually become popular now this
was one of the things that I actually
got to demo in person you can see they
had a little camera up top looking down
at these objects here and you were able
to ask questions about the objects it
would tell you stories about a T-Rex and
a duck you can see he just added this
dut here it would add the dut into the
story and all of this stuff worked in
real time uh this was me shooting from
my 360 camera so you can see me watching
here but I actually did get to demo this
and it seemed to work as they showed it
off in their demos on stage they also
showed off this feature in Notebook LM
where you can throw a whole bunch of
data in there throw in docs and
spreadsheets and your own audio messages
and all sorts of stuff into this
notebook LM and then it would actually
create this like interactive podcast for
you where it would explain whatever
information you dumped into it and you
can kind of cut it off and ask questions
of it it was really really cool they
showed off imagine 3 which is their new
text to image model that's getting
really really impressive these images
are looking much more realistic than the
previous generations of imagine so
here's some more examples these are
really approaching mid journey level of
realism in my opinion but what was even
cooler than imagine 3 was when they
showed off Veo their new video
generation model which looks like it
wants to compete with Sora I don't feel
like the quality of the demos they
showed are quite up to the same par as
what we see out of Sora but this is like
the second best video generation model
we've seen so far behind Sora it creates
videos in 1080P and from my
understanding it can actually generate
videos over 1 minute long and they even
managed to tap Donald Glover to make
videos using this new Veo video model
and we'll get to see the film that he
produced using this sometime soon now
while Sora we have absolutely no clue
when they're going to make it available
Veo is actually going to start letting
people use it they actually have a
waiting list over on the AI Test Kitchen
website and apparently are going to
start letting people actually use this
one Sora who knows we may not even see
it this year in our own hands the Google
search got some upgrades as well using
AI where now they claim that Google will
do the Googling for you they gave this
example of multi-step reasoning
capabilities where they were able to ask
a question to Google like find the best
yoga or Pilates studios in Boston and
show details on their intro offers and
walking time from Beacon Hill and it
would take in that entire question SL
prompt into Google do a search and
answer all of that for you another
really cool thing they showed off was in
Gmail you'll be able to have this little
chat box and ask questions like catch me
up on emails from Maywood Park
Elementary School and it will look
through all of your emails related to
school and give you a nice recap so you
don't have to skim through tons of
emails you get it all in one place
they're also going to have that in
Google meet and pretty much the entire
Google Suite of tools are going to get
this little sidebar here they even said
later this year in Labs you can ask a
Gemini to automatically organize email
attachments in Drive generate a sheet
with the data and then analyze it for
you they also showed off ask photos
which was a really cool feature where it
knows the context of all the photos and
you can say when did my daughter learn
to swim and it would look through all
your photos and find times where she was
swimming and tell you when she learned
to swim or remind me what themes we've
had for Lena's birthday parties here are
the themes for Lena's last four
birthdays third a princess celebration
fourth Under the Sea fifth and sixth
magical unicorn parties it found all
that information by just looking at the
photos one really really interesting
moment that the crowd kind of went crazy
for was when they showed this feature
where when somebody's trying to scam you
on a phone call not in text message on
an actual phone call it will try to
detect that and warn you that you might
be getting scammed right now check this
out let's say I get rudely interrupted
by an unknown caller right in the middle
of my
presentation hello hi I'm calling from
safe Mor Bank Security Department am I
speaking to Dave uh yeah this is Dave
kind of in the middle of something we've
detected some suspicious activity on
your account it appears someone is
trying to make unauthorized charges uh
oh yeah what what kind of charges I
can't give you specifics over the phone
but to protect your account I'm going to
help you transfer your money to a secure
account we've set up for
you and look at this my phone gives me a
warning that this call might be
astounding Gemini Nano alerts me the
second it detects suspicious activity
like a bank asking me to move my money
to keep it safe and everything happens
right on my phone so the audio
processing stays completely private to
me and I my device we're currently
testing this feature and we'll have more
updates to share later this summer now
there were a ton more announcements
again I shared an article that had a
hundred different announcements it was
just announcement overload and again I
also made a whole dedicated video just
talking about the announcements and my
feelings around those announcements so
if you do want to go deeper and learn
even more about what happened at Google
IO definitely check out that video as
well I think Joshua Scala here sort of
summed up the two events well the
difference between Google and open AI
open AI demos cool stuff and I know
where to go try it Google demos 50
things with all sorts of names random
places to go use them vague release
dates and random permission walls I need
to pay apply or move country for I'm
totally lost and thus disinterested now
I don't totally agree with that I am
super interested in what Google's
releasing I think they showed off some
really really cool stuff I also met a
lot of the people that were developing
these products a lot of the product
managers and the leads on some of these
things and the people building this
stuff were really excited about it they
were really passionate about what they
were building when you go to events like
this and I did mention this in my other
video when you go to events like this
you get a better concept of the humanity
Underneath It All Google is a big sort
of faceless Corporation but you go to an
event like this and you meet all the
people building this stuff and they're
just as interested in Ai and progressing
this stuff forward and just as nerdy and
passionate and really excited about the
work they're putting in and so I know a
lot of people are rooting against Google
but then when you meet the people that
are building this stuff and they're
excited you don't want to root against
them they're building this stuff that's
really cool that they're passionate
about and Google gives them a place to
go and try this out and build it but I
do agree that Google has some issues
with the naming conventions I even
talked to one person at Google who
admitted that it's sort of an inside
joke at Google that the naming of all
this stuff is confusing and and hard to
keep track of they also do have a
tendency to make big announcements and
say it's coming soon and then take
forever to release it I mean that's just
the reality but that does not diminish
what they're showing the tech that they
showed off is really cool we just want
to get our hands on it all right let's
move on past the open Ai and Google News
because again I've talked about it a lot
in this video already and I made two
other videos about those two events
already but there was some other albeit
more minor updates from some other
companies and I'll just quickly run
through those for you now as well
starting with anthropic anthropic hired
Instagram's co-founder as its head of
product Mike creger who was one of the
co-founders at Instagram and also the
co-founder of the artifact app a news
app that used AI to help surface
relevant news for you joined anthropic
as the company's Chief product officer
it sounds like his role at the company
is to engineer good user interfaces and
user experiences to get more people
excited and talking about and use it
anthropics tools like Claud and while
we're on the topic of anthropic they
just released a new prompt generator you
can now generate production ready
prompts in anthropic console describe
what you want to achieve and Claud will
use prompt engineering techniques like
Chain of Thought reasoning to create
more effective precise and reliable
prompts so if you go to console.
anthropic dcom you can see a little menu
here including one that says generate a
prompt if we click generate a prompt we
can give it some minor information about
what we want the prompt to achieve let's
say come up with several ideas for a
YouTube title for a video recapping a
lot of news in the AI world for the past
week let's generate the prompt and see
what it comes up with and you can see it
generated a pretty in-depth prompt here
your task is to come up with several
engaging YouTube video title ideas for a
video recapping recent developments in
the world of artificial intelligence
I'll provide you with a summary of the
recent AI news and you will use that to
brainstorm title ideas here's the
summary gives me a little area to plug
in a summary here first brainstorm after
brainstorming select three top titles
remember the goals to come up with
titles that will entice people to click
Etc so it took my short one sentence
prompt and built it into a really dialed
in prompt engineer optimized prompt to
use inside of clad a few weeks ago I
talked about a new AI company called
hum. a company that actually can detect
the emotion and feelings in your voice
when you speak like whether or not
you're happy or angry or excited things
like that well Hume just released a new
tool called chatter an interactive
podcast experience and basically this is
a podcast that you can sort of steer the
direction of it's hard to explain so
I'll just show you if I click Start here
it will just kind of start telling me
the news hello welcome to chatter an
interactive News podcast the date is May
16th 2024 today's top headlines are
absolutely riveting hey I'm going to
stop you real quick instead of talking
about overall news can you share just
the interesting news from the AI world
all right let's dive into the
fascinating world of AI today there's
some thrilling news from this ever
evolving field first up Nvidia continues
to dominate the AI scene their
processors are in high demand for cloud
computing driving significant AI growth
what were the big announcements that
came out from Google and open AI this
week oh yes all right let's dive into
the latest from Google and open AI
starting with Google they've made some
roue waves at their recent Google is O
event they introduced The Gemini AI
models which are set to revolutionize
how we interact with AI these models are
designed to be more intuitive and
efficient well I think you get the idea
you can sit there and have a
conversation with it and it's like
listening to the radio but you can sort
of inject your own questions it will
stop talking and then respond to
basically what you were asking it to
respond to it's pretty cool again that's
called chatter and you can find it over
at
chat. at the end of last week this clip
kind of went viral from Whitney wolf
herd who is the founder of Bumble and
she talked about what the future of
dating could look like specifically what
was interesting was the part where she
talked about how you'll have like an AI
dating concierge and your dating
concierge will go and date other
people's AI concierge and the AI dating
each other will decide if the two real
people are compatible so check this out
if you want to get really out there
there is a world where your dating
concierge could go and date for you with
other dating concierge uh uh no no no
trly and then you don't have to talk to
600 people it will scan all of San
Francisco for you and say these are the
three people you really autom
interesting I'm not in the dating pool
so I will probably not ever experience
this but I'm pretty dang sure that is an
episode of black mirr I think I saw that
one this week we learned that meta is
exploring AI assisted earphones with
cameras now there's not a lot of news
about this this seems to be kind of
early speculation but it sounds like
they're exploring making like airpod
type earbuds that have cameras on the
little tip of the earbuds so that they
can see what's going on and you know
imagine like the rabbit or the AI pen or
even the meta Rayband glasses that you
wear well this sounds like it could have
potentially do the same thing where you
can hear everything it's telling you but
it can also see what's going on around
you maybe it'll record video maybe the
camera is just to see what's going on to
give context to the llm I don't know
there's not a lot of details about this
yet this seems to be kind of a rumored
thing but seemed interesting so I
thought I'd share and finally this is
just the beginning of event season next
week is Microsoft's build event which is
sort of the equivalent of Google IO but
Microsoft's version and Microsoft tends
to do similar to Google where they do
these events instead of a 30-minute
keynote like open AI they do a 2hour
keynote and just bombard you with
announcements so that's kind of what I'm
expecting from Microsoft also now that
we have GPT 40 we also know that
Microsoft Works really closely with open
AI Microsoft is open ai's biggest
investor I also kind of have a feeling a
lot of the announcements are going to be
this product now has GPT 40 in it this
product now has GPT 40 in it it's
probably going to be a lot of co-pilot
now works with GPT 40 co-pilot now has a
voice where you can just chat directly
with it I think we'll probably see some
of what we heard from open AI being
rolled into to a lot of Microsoft
products this is another event that I'm
going to be at so I'm going to be doing
my best to keep you informed on what I
learn at this event you'll probably see
more videos of me recording from a hotel
room but I want to learn as much as I
can meet as many people as I can at
these events and turn around and share
what I'm seeing and what I'm learning
and what I think is really cool that's
in the pipeline so that's happening next
week so next week's news will probably
be mostly dominated by Microsoft and
then in June we have a Cisco event where
they're going to be talking about Ai and
cyber security we have a Qualcomm event
where they're probably going to be
talking a lot about their Snapdragon
chips and doing more AI processing on
mobile devices and then we also have an
Apple event in June their worldwide
developer conference where they're
expected to be unveiling a whole bunch
of AI features there's a lot of events
coming up I'm going to be at most of
them I'm not going to be at the Apple
event but I'll be at the Cisco event
I'll be at the Microsoft event and I
will be at the Qualcomm event sharing
whatever I learned from all of those
events in their own respective videos so
super excited we're just getting started
the AI sort of hype time of the year is
ramping up and this is this is what I'm
here for I love this time of year I love
it when all these announcements are
coming out and I get to try new tools
and make videos and tell you about
what's coming out there is so much
you're probably going to see a lot of
more videos from me over the next couple
weeks because there's going to be a a
lot to talk about but that's what I got
for you this week again it was mostly
Google and open AI with a few other
little bits and Bobs sprinkled in there
next week it's probably going to be
mostly Microsoft I'm here for it it's
going to be exciting if you want to stay
in the loop with all the latest AI news
on a daily basis make sure you check out
futur tools. I do keep the news up to
date on mostly a daily basis when I am
traveling it updates a little bit slower
because I am still approving every piece
of news that goes on here cuz I want it
to stay really high signal not a lot of
noise on this news page I also curate
all of the coolest AI tools that I come
across with some really easy filtering
to just get down to the tools that you
need for whatever it is you're trying to
do and we've got a free newsletter you
join the free newsletter and I will Loop
you in every week with just the coolest
AI tools that I come across and just the
most important news that I think you
need to know like we're talking three to
five news articles tops per week so
you're not getting bombarded with
everything it's just the most impact F
most important stuff check that out also
I am giving away a brand new insta 360
X4 this thing just came out this is like
a week old camera brand new from insta
360 you can be entered to win one
totally for free all you have to do is
make sure you're subscribed to this
YouTube channel and you join our free
newsletter as long as you're subscribed
in both of those places you are entered
to win this camera and I'm going to be
doing giveaways like this every single
month now I'm in a lucky position where
a lot of these companies send me these
gadgets to play around with and try and
make videos with and I thought it would
be fun to turn around and turn some of
this stuff into competitions and
giveaways and Raffles and things like
that so make sure you're on the
newsletter and make sure you're
subscribed to this channel you'll be
iner to win and you'll also be one of
the most looped in people on what's
going on in the world of AI because I'll
keep you looped in thank you so much for
tuning in really really appreciate you
spending the time with me today nering
out on everything Ai and I will see you
in the next video really appreciate you
bye-bye
Ver más vídeos relacionados
Why OpenAI's Announcement Was A Bigger Deal Than People Think
These AI Use Cases Will Affect Everyone You Know
Всё о новой нейросети GPT-4o за 7 минут!
OpenAI presenta ChatGPT-4 OMNI (GPT-4o): GPT ORA SEMBRA AVERE EMOZIONI!
AI News: The AI Arms Race is Getting Insane!
GPT-4o is WAY More Powerful than Open AI is Telling us...
5.0 / 5 (0 votes)