Was GPT-5 Underwhelming? OpenAI Co-founder Leaves, Figure02 Arrives, Character.AI Gutted, GPT-5 2025
Summary
TLDRThe transcript discusses the current state of AI development, highlighting delays in GPT-5's release and leadership changes at OpenAI, including Greg Brockman's leave and John Schulman's departure to Anthropic. The narrative mentions other AI advancements, like Meta's Llama models and Google's Gemini 1.5 Pro, suggesting OpenAI may be falling behind. It also covers upcoming events like OpenAI's Dev Day and the ongoing lawsuit from Elon Musk against Sam Altman. Additionally, the transcript touches on humanoid robots using OpenAI's technology and the broader AI landscape's shifting dynamics.
Takeaways
- ๐ The GPT 5 release date is uncertain, with Open AI leadership experiencing changes and departures.
- ๐ง Greg Brockman, co-founder of Open AI, is taking a leave of absence, highlighting the ongoing challenges in achieving safe AGI.
- ๐ข There is a trend of departures among Open AI co-founders, including John Scharff who left for Anthropic, signaling potential alignment issues.
- ๐ Open AI may be lagging in raw intelligence compared to models like Llama with 45 billion parameters, suggesting a need for significant improvement in their next model.
- ๐ฎ The anticipation for GPT 5 is high, but its release might be delayed beyond the expected timeframe, possibly due to strategic planning around developer events.
- ๐ค The development of agile, autonomous humanoid robots like Fig2 indicates a potential shift towards practical applications of AI beyond just model development.
- ๐ Elon Musk's lawsuit against Sam Altman and Open AI raises questions about the original intentions and current direction of Open AI.
- ๐ The debate over who will control the future of AI is intensifying, with concerns about the concentration of power in a few large companies.
- ๐ The LM CIS chatbot Arena leaderboard shows a competitive landscape with models like Gemini 1.5 Pro and Claude 3.5 Sonic outperforming others.
- ๐ There is a growing trend of smaller AGI labs being overshadowed by larger entities with more resources, such as Google acquiring talent from Character AI.
- ๐ The importance of a data flywheel effect is emphasized, where more data leads to better AI, which in turn can generate more data.
Q & A
What is the significance of the departure of John Scharff from OpenAI to Anthropic?
-John Scharff's departure to Anthropic is significant because it indicates a trend of departures among co-founders at OpenAI, and it suggests that he may not have been working with people who were deeply engaged with the alignment work he was focused on, which is an attempt to align machine values with human values.
What does the term 'alignment' refer to in the context of AI?
-In the context of AI, 'alignment' refers to the effort to align machine values with human values, ensuring that AI systems behave in ways that are beneficial and safe for humans.
Why is the release of GPT-5 potentially delayed beyond expectations?
-The release of GPT-5 could be delayed because OpenAI has announced events focusing on advancements in APIs and dev tools, which implies that they are not planning to release a new model immediately after these events, suggesting a release date more likely at the end of the year or later.
What is the advanced voice mode mentioned in the script, and how is it related to OpenAI's models?
-The advanced voice mode is a feature that makes the voice output of AI models incredibly lifelike. It is related to OpenAI's models as it could potentially be a part of their future offerings, enhancing user interaction with their AI systems.
What is the controversy surrounding Elon Musk's lawsuit against Sam Altman and OpenAI?
-Elon Musk is suing Sam Altman and OpenAI, accusing them of lying and deceit, claiming that Altman was motivated by greed and that Musk wanted OpenAI to be more open to compete with Google. The lawsuit describes the original OpenAI as a 'spurious venture' and uses strong language to emphasize the accusations.
How does the script suggest that smaller AI labs like Character AI are struggling to compete with larger labs?
-The script suggests that smaller AI labs like Character AI are struggling because they are being overshadowed by the scale of data and compute resources that larger labs like Meta and Google have, which is making it difficult for them to develop models that are as intelligent as those of the larger labs.
What is the significance of the LM Cis Chatbot Arena leaderboard for AI models?
-The LM Cis Chatbot Arena leaderboard is significant as it provides a ranking of AI models based on their performance in various tasks, offering a comparative measure of their capabilities and helping to identify trends and strengths among different models.
Why might the new Gemini 1.5 Pro version be considered a more justified leader on the LM Cis Chatbot Arena leaderboard compared to GPT-40 Mini?
-The new Gemini 1.5 Pro version might be considered a more justified leader because it performs slightly worse than 3.5 Sonic but far better than other models, suggesting a more balanced and comprehensive performance across various tasks.
What is the potential impact of the advanced voice mode on the adoption of AI models by the public?
-The advanced voice mode could significantly increase the adoption of AI models by the public because it makes interactions with AI systems more natural and lifelike, potentially leading to hundreds of millions of people using it once it becomes available.
What does the script suggest about the future of AI control and competition among different companies?
-The script suggests that the future of AI control is becoming increasingly concentrated among a few large companies, with OpenAI, Meta, and Google leading the way. It raises questions about who will control the future of AI and the implications of these companies' dominance in the field.
Outlines
๐ Open AI Leadership and Departures
The script discusses the shifting landscape of AI leadership, with Open AI facing internal changes as Greg Brockman takes a leave of absence and several key figures depart, including John Schan, the former head of alignment. The departures suggest a lack of alignment with the company's direction, especially concerning the critical issue of AI alignment with human values. There is speculation about the capabilities of Open AI's next frontier model, which may have already been trained, and the implications of these departures on the company's progress towards AGI. The script also touches on the lawsuit by Elon Musk against Sam Altman and Open AI, accusing them of deceit and greed, and the broader debate on who will control the future of AI.
๐ AI Competition and Model Advancements
This paragraph delves into the competitive AI landscape, highlighting the strides made by companies like Meta with their Llama model and the challenges faced by Open AI to keep pace. It discusses the potential release of advanced models like Claude 3.5 and the significance of open-source competition encouraged by the White House. The script also points out the struggles of smaller AI labs like Character AI, which have been overshadowed by tech giants like Google. The discussion includes the performance of various AI models on benchmarks and leaderboards, emphasizing the difficulty in measuring AI intelligence and the nuances of model capabilities.
๐ค Integration of AI with Robotics and Future Prospects
The final paragraph focuses on the integration of AI with robotics, particularly the adoption of Open AI's models by Figure O2 humanoid robots. It describes the capabilities of these robots, including advanced voice interaction and physical strength equivalent to humans. The script speculates on the potential for a data flywheel effect as these robots perform tasks autonomously and self-correct. It also mentions the use of Weights & Biases for tracking machine learning experiments and the importance of iterating on LLM applications. The paragraph concludes by reflecting on the shifting dynamics in the AI field and expressing gratitude to the audience for their engagement.
Mindmap
Keywords
๐กAGI (Artificial General Intelligence)
๐กOpenAI
๐กAlignment
๐กGPT (Generative Pre-trained Transformer)
๐กCo-founder Departures
๐กAdvanced Voice Mode
๐กLM CIS Chatbot Arena
๐กWeights and Biases
๐กData Flywheel
๐กFig2
๐กElon Musk
Highlights
GPT 5 release date is uncertain, with Open AI leadership facing challenges.
Greg Brockman, co-founder of Open AI, is taking a leave of absence until the end of the year.
John Scharf, another co-founder, has left Open AI for Anthropic, indicating a lack of alignment with the company's direction.
There is a trend of departures among co-founders at Open AI, raising concerns about the company's future.
Open AI has begun training its next Frontier Model, but details on its capabilities remain unknown.
Open AI's Dev Day in October suggests that GPT 5 will not be released before the end of November 2024.
Elon Musk is suing Sam Altman and Open AI, alleging deceit and a lack of openness in the company's operations.
Meta and Google are leading in AI development, with smaller labs struggling to compete.
Character AI, a smaller AGI lab, has been overshadowed by Google's acquisition of key talent and IP.
The debate on AI intelligence suggests that scale in data and compute is more important than obscure tricks or arcane knowledge.
LM CIS chatbot Arena leaderboard shows new models like Gemini 1.5 Pro taking the lead in performance.
Open AI's advanced voice mode is incredibly lifelike, suggesting potential for widespread adoption.
Figure O2 humanoid robots are using Open AI models, indicating a practical application of AI technology.
AI Labs are increasingly relying on tools like Weights & Biases to track machine learning experiments.
The shift in AI development suggests that control over the future of AI is becoming more contested.
The release of raw data by LMIS provides insights into model performance and their refusal to answer certain questions.
The new Gemini 1.5 Pro is being tested for its reasoning capabilities, with results expected to be published online.
AI development is moving towards more practical applications, such as autonomous humanoid robots, rather than just theoretical advancements.
Transcripts
as the GPT 5 release date sails further
into the Horizon open AI leadership
splinters meanwhile other AI Labs ship
incrementally smarter models and smaller
AGI efforts like character AI are
swallowed by the Google whale leaving in
my eyes just four companies remaining in
contention for having the most capable
models out there but if we get agile
autonomous humanoid robots like fig2
that could soon count to 50 in one
breath with chat gbt advanced voice
maybe many will take that as a win for
2024 but let's start with the leadership
of open Ai and Greg Brockman is going on
a leave of absence through to the end of
the year calling it his first time to
relax since co-founding the company he
goes on the mission is far from complete
though we still have a safe AGI to build
now of course any one person taking time
away might well be for person reasons
but we have the full-on departure to
anthropic of another one of the
co-founders John schan the reason that
he as the former head of alignment at
openai jump ship to anthropic I think is
given away in this sentence he said he
wants to do research alongside people
deeply engaged with the topics he's most
interested in now it is pretty hard to
read that as saying anything other than
he wasn't working with people who were
deeply engaged with the kind of
alignment work that he was working on
Just for those who don't know alignment
is this attempt to align machine values
with human values but it follows the
departure of the previous head of
alignment Yan Leica and before that the
previous co-head of alignment and
co-founder Ilia satova I'm definitely
starting to notice a trend of Departures
among co-founders at openai so I've
listed chat PT to count the former heads
of alignment at open AI 1 2 3 4 5 6 7 8
9 10 11 12 13 14 15 16 17 18 192 21 22
23 24 25 26 27 28 29 30 31 32 33 34
obviously I was somewhat joking there I
think there's been less than 10 former
heads of alignment but that advanced
voice mode does seem cool I mean even if
open AI models aren't actually getting
smarter and arguably with GPC 40 are
getting slightly Dumber that advanced
voice mode is incredibly lifelike and I
could see hundreds of millions of people
using it at least when it comes out
which seems to be on the Never Never and
here's some even more important context
open aai back in May said that they had
recently begun training its next
Frontier Model now here we are on August
the 6th it would almost certainly have
finished training by now so those key
players at openai would have a rough
sense for its capabilities all of these
departures after they've trained their
latest Frontier Model seems strange and
then we got this about open ai's
so-called Dev day which is starting in
October October and running through to
November I am sure it will be
fascinating but they released this
particular nugget while we know
developers are waiting for our next big
model which we shared has begun training
earlier this year in May these events
will focus on advancements in the API
and are Dev tools taking that at face
value would imply that quote GPT 5 will
not come before November 21st more
likely it means that gbt 5 wouldn't even
come before the end of the year because
why would you release a model just after
you've invited a load of devs to play
about with your tools now yes samman has
recently claimed in the Washington Post
that more advances will soon follow and
will usher in a decisive period in the
story of human society but in recent
months it would be hard to say that open
AI have produced much in the way of
decisive progress and of course all of
that comes as Elon Musk is suing yet
again Sam mman and open AI for what he
says is lying and pery the lawsuit calls
the original open AI a spurious Venture
the language used throughout this
86-page document is hardly subtle musk
claims that samman is doing a long con
his pery and deceit are of Shakespearean
proportions now it does go on and on but
the basic accusation is that Sam ultman
was motivated by greed and Elon Musk
just wanted to have something more open
to compete versus Google so I think musk
and Others May raise an eyebrow when
later in the article samman said making
sure open-source models are readily
available to developers in other nations
will further bolster our advantage
talking of the US and the question that
the article fundamentally raises is who
will control the future of AI well as of
today it looks less and less likely to
be Sam Morman open AI are certainly good
at productizing Ai and the advanced
voice mode as we saw is great search GPT
could make them some money and S is
coming out at some point presumably this
year and at least at the moment the
figuro 2 robot is using an open AI video
language model we'll get back to that in
just a moment but in terms of raw
intelligence open AI feel like they're
falling behind the Llama 3 45 billion
parameter model is already smarter than
GPT 40 and Zuckerberg has recently
committed to 10 times more computing
power to train llama 4 or to put it
another way the next open AI model would
have to be significantly better than GPT
40 just to catch up to the current
state-ofthe-art let alone the
state-ofthe-art when llama 4 comes out
and of course in the meantime even just
this year we might be getting Claude 3.5
Opus from anthropic or even Claude 4
simply put a year is a long time in Ai
and the debate has moved on even the
White House are now encouraging
open-source competition to the light of
open AI Sam Alman meanwhile is still
warning about people stealing key
intellectual property such as model
weights now do forgive me for pointing
out that one of the modules on my
corsera course is about that difference
between open source and open weights
super grateful of course for those 10
reviewers who have kindly left reviews
for this course but don't get me wrong
it's not like meta is having it all its
own way do you remember those Tom Brady
Paris Hilton chatbots that all the
youngsters were apparently going to be
using I think each celebrity was paid
something like $5 million for a few
hours of recordings well apparently
they're now being scrapped and none of
those AI chat Bots amassed a
particularly big following but nor are
things going particularly well for these
smaller AGI Labs like character AI their
product was or is an array of chat Bots
but they were also aiming at AGI and
training Their Own Foundation models
obviously the leaders of character AI
must have been somewhat disappointed by
those New Foundation models because
essentially they've been bought out or
hollowed out by Google not actually
buying a rival company but taking its
key talent and IP but at this point you
might be starting to notice somewhat of
a trend if the incrementally greater
intelligence of new models were down to
obscure tricks or Arcane knowledge then
you'd expect smaller Labs like character
AI to be doing as well as the biggest
Labs but if it was all about sheer scale
of data and compute you'd expect the
leaders to be increasingly well meta and
Google and that is more or less what
we're seeing though of course measuring
that intelligence is quite hard we do
have the LM CIS chatbot Arena
leaderboard in which the new version of
Gemini 1.5 Pro takes the lead at almost
1300 ELO but if you'll notice we have
GPT 40 mini coming third ahead indeed of
Claude 3.5 Sonic which in my own
Benchmark is far and a way ahead if we
just relied on these ELO rankings you'd
think well if open AI can come up with a
tiny model doing almost as good as the
rest they must be doing amazingly but
lmis recently did something great which
was release a batch of raw data showing
comparisons between the models and which
one won and I looked through the dozens
of examples and one Trend emerged claw
3.5 Sonic essentially refused more
requests than GPT 40 Mini even when both
models couldn't perform a task like
creating an image natively GPT 40 mini
gave it I guess more of a go it at least
describe the image that it would create
or in this example when the models were
asked a political question GPC 40 mini
gave a response whereas Claude 3.5 Sonic
just apologized and said it wouldn't
provide analysis now I have noticed that
myself that Claude 3.5 Sonic is more
sensitive than any other model but
that's not a sign of lacking
intelligence so to the the extent that
we're going to call language modeling
intelligence Claude 3.5 Sonic is far
more capable than GPC 40 mini but this
Rance to answer certain questions could
explain the leaderboard rankings now yes
to anyone following the channel I have
been testing the new version of Gemini
1.5 Pro on my simple bench the final
scores will be presented on a website
that I'm hoping to release before the
next video but in the early testing it
performs slightly worse than 3.5 Sonic
but far better than other models so in
that sense this lead B position could be
far more Justified at least than GPT 40
mini for those who haven't heard of my
new reasoning Benchmark humans score
over 90% quite easily whereas models
like the new Gemini 1.5 pro version
score around 25% it would however be
somewhat hypy to say that this is
another step toward AGI as one of the
co-founders of Google deepmind recently
said but whether you think that new
Gemini 1.5 Pro is the best or Claude 3.5
Sonic certainly more and more people are
now shifting their API spend away from
openai openai letting other labs have
the lead for a couple of weeks could be
a timing issue but a couple of months
just makes it seem like they don't have
a reply but as I mentioned at the start
at least open a eyes models are the ones
being chosen by figure O2 these humanoid
robots have an onboard mic and speaker
so hopefully you could chat to them like
you would the new openai advanced voice
mode in other words seamlessly with very
low latency Adcock the founder of figure
said that the default user interface to
our robot will be speech apparently the
robot can work for around 20 hours
straight which is even more than me
reading the latest AI papers its hands
have 16 degrees of freedom and
apparently human equivalent strength so
honestly even though it can speak back
to you you might not want to speak back
to it now apparently these fig2 robots
can perform certain tasks autonomously
and self-correct and that data flywheel
will be in effect and one might well say
that the argument that ubiquitous robot
assistance will arrive before artificial
general intelligence looks more
plausible than ever and speaking of a
data flywheel you may already know that
AI Labs including open AI have used
weights and biases this video sponsor to
track Frontier machine learning
experiments but what you might not know
is that weit and bu now have a weave a
lightweight tour kit to confidently
iterate on llm applications they also
produce free prompt and llm agent
courses on their website and if you
didn't know that you can let them know
that you came from me by using my
customized link and the link is in the
description so in short the vibe is
Shifting but what isn't changing is my
gratitude for you watching all the way
to the end if you're Keen to carry on
the conversation with me personally I'd
love to see you over on AI ins iders on
patreon but to everyone watching have a
wonderful day
Browse More Related Video
AI News: This Was an INSANE Week in AI!
HUGE AI NEWS : MAJOR BREAKTHROUGH!, 2x Faster Inference Than GROQ, 3 NEW GEMINI Models!
Massive AI News: Google TAKES THE LEAD! LLama 4 Details Revealed , Humanoid Robots Get Better
Huge AI NEWS : OpenAI Gives GPT-5 To US Government, Stunning New Humanoid Robot, Metas New Model..
GPT-5 SOON, AI-to-AI Payments Using Crypto, xAI GPU Cluster is Live, 1,000 Agent Simulation
20 Surprising Things You MISSED From SAM Altman's New Interview (Q-Star,GPT-5,AGI)
5.0 / 5 (0 votes)