March 7, 2025 • 8 mins
In this week's monologue, Ed Zitron goes through the underwhelming release of OpenAI's GPT-4.5 - and how OpenAI's rumored new pricing structure suggests this company is in trouble.
Here's the article cited from The Information: https://www.theinformation.com/articles/openai-plots-charging-20-000-a-month-for-phd-level-agents?rc=kz8jh3
---
LINKS: https://www.tinyurl.com/betterofflinelinks
Newsletter: https://www.wheresyoured.at/
Reddit: https://www.reddit.com/r/BetterOffline/
Discord: chat.wheresyoured.at
Ed's Socials:
https://www.instagram.com/edzitron
https://bsky.app/profile/edzitron.com
https://www.threads.net/@edzitron
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:02):
Zone Media. Hello you, it's your better offline monologue and
I'm your host ed ZiT trum. Now next week I'm
going to have a two part of that digs into
Microsoft Stata senter pull back an open ai is shaking
you funding situation. But this week's monologue focuses an open
(00:25):
AI's new model, GPT four point five. You may be
wondering what it does differently to GPT four oh, or
claudes on at three point seven, or any number of
other large language models, And if I'm honest, I have
absolutely no idea. Thankfully neither does Sam Ortman, CEO of
open Ai, who said, and I quote, the GPT four
point five was the first model that feels like talking
(00:47):
to a thoughtful person to him, which makes me wonder
what the other models have been like. So I went
back and look. So compare this to the launch of
GPT four oh, which Altman called open AI's best model ever,
saying that it was fast, smart, natively multimodal, referring to
the ability to accept text as well as audio and
(01:07):
video and photos as well, and available to all chat
GPT users, including on their free plan, adding that it
was a very good model, especially at Coding. By contrast,
Alltman summarized GPT four point five as a giant, expensive model,
one that required hundreds of thousands of GPUs to launch
beyond chat GPT pro. It started there. It's still not
(01:28):
as I record this out to plus users or free users. Now.
GPT pro of course is open AI is two hundred
dollars a month subscription, and it's unclear when plus, which
is the twenty buck a month one will get it,
but apparently it's in the next few days. Aortman also
added that GPT four point five isn't a reasoning model
and won't crush benchmarks on account of it being a
(01:50):
different kind of intelligence that has and all of these
are quotes magic to it that Sam Mortman had not
felt before. Yeah, just you know, shit's not doing well
when you have to just be like it's magic, It's
it's literally magic. I made magic. Now what does the
magic do? I'm really not sure. In fact, it's pretty
(02:13):
difficult to find exactly what it is that GPT four
point five does differently, or what it's good at, or
indeed really anything about him. BENJ. Edwards over At asked
Tenneker had one developer called it a lemon. GPT four
point five costs an incredible seventy five dollars per million
input tokens prompts and data pushed into a model, and
one hundred and fifty dollars per million output tokens is
(02:33):
in the thing it creates. A token is like zero
point seven. I think one token is maybe three words.
Someone will get up my ass for this. Nevertheless, this
seems like a lot. It isn't when you're running a company.
And by the way, this is roughly three thousand percent
more expensive for input tokens and fifteen hundred percent more
expensive for output tokens than GPT four to zero for
results that open ai co founder Andredge Carpathy described as
(02:57):
a little bit better and awesome, but also not exactly
in ways that are trivial to point to. That translates
to it's a little bit better, but I can't really
tell you why. And yes, you're gonna hear me say
something similar in next week's episode, because the larger picture
for open ai right now is pretty fucking dire considering
their main backer, soft Bank, has to borrow billions of
(03:18):
dollars to fund them. Nevertheless, back to four point five
Since launch, which was for some reason, on the day
that Sam Ortman's child was being born in the hospital,
He's been posting some really weird shits since, though. A
few days after launch, Aortman claimed that GPT four point
five was the first time people had been emailing with
such passion, asking the open AI promised never to stop
(03:41):
offering a specific model or even replace it with an update,
at which point I assume everybody in the room started
clapping and they saluted Sam Mortman and said, thank you,
sir for making this happen. And by the way, what
I'm suggesting is that no one's ever done this, or
like one freak did, or maybe Aortman emailed it to himself,
who just shut up? Just your company burns five billion
(04:04):
dollars a year and the best you've got is this
warmed up dog shit about people marine todding you over
your model and never taking it away. Has opening I
ever even taken away a model? Jesus fucking these companies anyway.
A few days later, Altman posted a conversation where he
asked GPT four point five if it believed it was real,
(04:24):
leading to a series of bullet points with things like
what do we mean by real only for GPT four
point five, saying that it believed that it was not
an independent consciousness, but rather a structured experience happening within
your consciousness, referring to Sam Altman, which is the kind
of shit that's only impressive if you're an imbecile or
so stoned you've texted date of your friends the question
what if the joker was Batman? And by the way,
(04:45):
the answer to that is called The Batman Who Laughs
and it's one of the worst comics ever written. If
you want to talk to me about DC Metal, please
email me. It's easy. That's e z or z. If
you're Canadian or British at Better Offline dot com I
a really if you are working for DC Comics right
now and you had anything to do with Death Metal
or The Batman Who Laughs, you and I have a grievance.
(05:07):
You and I need to talk. Sorry what this is
a tech pop us right back to open AI. More worryingly,
Samultman posted an idea for paid plans where your twenty
dollars plus subscription converts the credits you can use across
speech just like Deep Research O one. GPT four point five, Sora,
and so on, with no fixed limits per feature, and
you choose what you want. If you run out of credits,
(05:29):
you can buy more. This, to be clear, is an
attempt to raise prices without actually raising them by attempting
to limit usage of open AI's more expensive models. Chat,
GPT plus and other subscriptions give you a limit, for example,
a limit of eighty messages every three hours on GPT
four to oh, but using one doesn't limit your use
of other products. Here, open ai is trying to create
(05:49):
a rent seeking model where power users have to pay
for more credits if they want to use, say open
ai is more expensive models like Sora and O one,
and I imagine any situation like this will be one
where they hope that people simply won't use their credits
or overuse them and have to pay for top ups.
This is, of course all theoretical, but it heavily suggests
that open ai is getting desperate. And now the information
(06:10):
is reporting that open ai executives have told some investors
that they will be charging two thousand dollars per month
for their low end agent product. And yes that's a
quote sold to And again I quote high income knowledge
workers with supposed mid tier agents for software development costing
possibly ten thousand dollars a month, with supposed PhD level
(06:31):
research agents costing twenty thousand dollars a month, and I
will tell you the PhDs I know would probably do
it for half, and they'd even work for an annoying
asshole like Sam Altman. Now you may wonder what any
of these things do, and the answer is that neither
I nor the Information know. As of right now. The
only operational or agent, open ai has his operator, open
AI's agent that sometimes successfully uses a web browser search
(06:53):
for something in minutes, which would usually take you seconds.
The Information attempted to suggest that the two thousand a
month agent would be some sort of thing that could
sort through and rank sales lead But I'm sorry, do
I really have to read this shit with a straight face?
Twenty thousand dollars for a PhD level agent? What the
fuck does that mean? What would it do? Why do
these companies? I get emails every week having to justify
(07:16):
my fucking cynicism, But these shittheads, they're allowed to just
make up stuff and it leak it to the information
The Information publishes. It were all meant to be impressed.
What the fucking what the fuck I'm allowed to rent
on these? They're allowing me to rant on these. It's
just it sickens me. I have had this week at
least five people email me and be like, well ed,
(07:37):
what would it take to change your mind about this stuff?
Why do I have to fucking do it? Why do
I The multi billion dollar companies do a dogshit job
of actually explaining this stuff or selling it. They lose
billions of dollars. But I'm the guy who has to
justify myself. Oh well, I'll keep doing it. Nevertheless, NES's
old At the bottom of this article's a far more
obvious payale horse. Open ai is planning to charge twenty
(07:59):
percent to thirty percent of pro customers the two hundred
dollars a month subscription that loses the money every time,
a higher price because of how many research queries they're
doing with Altmann, according to the information, suggesting some sort
of hey guessoir a la carte or pay as you
go approach. I want to be clear about something. This
is not a company that's cooking. This is not a
(08:21):
company that's worked out anything. Open Aiye is unprofitable, unsustainable,
and deeply, deeply lost. These are the actions of a
desperate company run by a desperate man. You have only
Sam Morman had a thoughtful friend to talk about all
these problems too,
Zone Media. Hello you, it's your better offline monologue and
I'm your host ed ZiT trum. Now next week I'm
going to have a two part of that digs into
Microsoft Stata senter pull back an open ai is shaking
you funding situation. But this week's monologue focuses an open
(00:25):
AI's new model, GPT four point five. You may be
wondering what it does differently to GPT four oh, or
claudes on at three point seven, or any number of
other large language models, And if I'm honest, I have
absolutely no idea. Thankfully neither does Sam Ortman, CEO of
open Ai, who said, and I quote, the GPT four
point five was the first model that feels like talking
(00:47):
to a thoughtful person to him, which makes me wonder
what the other models have been like. So I went
back and look. So compare this to the launch of
GPT four oh, which Altman called open AI's best model ever,
saying that it was fast, smart, natively multimodal, referring to
the ability to accept text as well as audio and
(01:07):
video and photos as well, and available to all chat
GPT users, including on their free plan, adding that it
was a very good model, especially at Coding. By contrast,
Alltman summarized GPT four point five as a giant, expensive model,
one that required hundreds of thousands of GPUs to launch
beyond chat GPT pro. It started there. It's still not
(01:28):
as I record this out to plus users or free users. Now.
GPT pro of course is open AI is two hundred
dollars a month subscription, and it's unclear when plus, which
is the twenty buck a month one will get it,
but apparently it's in the next few days. Aortman also
added that GPT four point five isn't a reasoning model
and won't crush benchmarks on account of it being a
(01:50):
different kind of intelligence that has and all of these
are quotes magic to it that Sam Mortman had not
felt before. Yeah, just you know, shit's not doing well
when you have to just be like it's magic, It's
it's literally magic. I made magic. Now what does the
magic do? I'm really not sure. In fact, it's pretty
(02:13):
difficult to find exactly what it is that GPT four
point five does differently, or what it's good at, or
indeed really anything about him. BENJ. Edwards over At asked
Tenneker had one developer called it a lemon. GPT four
point five costs an incredible seventy five dollars per million
input tokens prompts and data pushed into a model, and
one hundred and fifty dollars per million output tokens is
(02:33):
in the thing it creates. A token is like zero
point seven. I think one token is maybe three words.
Someone will get up my ass for this. Nevertheless, this
seems like a lot. It isn't when you're running a company.
And by the way, this is roughly three thousand percent
more expensive for input tokens and fifteen hundred percent more
expensive for output tokens than GPT four to zero for
results that open ai co founder Andredge Carpathy described as
(02:57):
a little bit better and awesome, but also not exactly
in ways that are trivial to point to. That translates
to it's a little bit better, but I can't really
tell you why. And yes, you're gonna hear me say
something similar in next week's episode, because the larger picture
for open ai right now is pretty fucking dire considering
their main backer, soft Bank, has to borrow billions of
(03:18):
dollars to fund them. Nevertheless, back to four point five
Since launch, which was for some reason, on the day
that Sam Ortman's child was being born in the hospital,
He's been posting some really weird shits since, though. A
few days after launch, Aortman claimed that GPT four point
five was the first time people had been emailing with
such passion, asking the open AI promised never to stop
(03:41):
offering a specific model or even replace it with an update,
at which point I assume everybody in the room started
clapping and they saluted Sam Mortman and said, thank you,
sir for making this happen. And by the way, what
I'm suggesting is that no one's ever done this, or
like one freak did, or maybe Aortman emailed it to himself,
who just shut up? Just your company burns five billion
(04:04):
dollars a year and the best you've got is this
warmed up dog shit about people marine todding you over
your model and never taking it away. Has opening I
ever even taken away a model? Jesus fucking these companies anyway.
A few days later, Altman posted a conversation where he
asked GPT four point five if it believed it was real,
(04:24):
leading to a series of bullet points with things like
what do we mean by real only for GPT four
point five, saying that it believed that it was not
an independent consciousness, but rather a structured experience happening within
your consciousness, referring to Sam Altman, which is the kind
of shit that's only impressive if you're an imbecile or
so stoned you've texted date of your friends the question
what if the joker was Batman? And by the way,
(04:45):
the answer to that is called The Batman Who Laughs
and it's one of the worst comics ever written. If
you want to talk to me about DC Metal, please
email me. It's easy. That's e z or z. If
you're Canadian or British at Better Offline dot com I
a really if you are working for DC Comics right
now and you had anything to do with Death Metal
or The Batman Who Laughs, you and I have a grievance.
(05:07):
You and I need to talk. Sorry what this is
a tech pop us right back to open AI. More worryingly,
Samultman posted an idea for paid plans where your twenty
dollars plus subscription converts the credits you can use across
speech just like Deep Research O one. GPT four point five, Sora,
and so on, with no fixed limits per feature, and
you choose what you want. If you run out of credits,
(05:29):
you can buy more. This, to be clear, is an
attempt to raise prices without actually raising them by attempting
to limit usage of open AI's more expensive models. Chat,
GPT plus and other subscriptions give you a limit, for example,
a limit of eighty messages every three hours on GPT
four to oh, but using one doesn't limit your use
of other products. Here, open ai is trying to create
(05:49):
a rent seeking model where power users have to pay
for more credits if they want to use, say open
ai is more expensive models like Sora and O one,
and I imagine any situation like this will be one
where they hope that people simply won't use their credits
or overuse them and have to pay for top ups.
This is, of course all theoretical, but it heavily suggests
that open ai is getting desperate. And now the information
(06:10):
is reporting that open ai executives have told some investors
that they will be charging two thousand dollars per month
for their low end agent product. And yes that's a
quote sold to And again I quote high income knowledge
workers with supposed mid tier agents for software development costing
possibly ten thousand dollars a month, with supposed PhD level
(06:31):
research agents costing twenty thousand dollars a month, and I
will tell you the PhDs I know would probably do
it for half, and they'd even work for an annoying
asshole like Sam Altman. Now you may wonder what any
of these things do, and the answer is that neither
I nor the Information know. As of right now. The
only operational or agent, open ai has his operator, open
AI's agent that sometimes successfully uses a web browser search
(06:53):
for something in minutes, which would usually take you seconds.
The Information attempted to suggest that the two thousand a
month agent would be some sort of thing that could
sort through and rank sales lead But I'm sorry, do
I really have to read this shit with a straight face?
Twenty thousand dollars for a PhD level agent? What the
fuck does that mean? What would it do? Why do
these companies? I get emails every week having to justify
(07:16):
my fucking cynicism, But these shittheads, they're allowed to just
make up stuff and it leak it to the information
The Information publishes. It were all meant to be impressed.
What the fucking what the fuck I'm allowed to rent
on these? They're allowing me to rant on these. It's
just it sickens me. I have had this week at
least five people email me and be like, well ed,
(07:37):
what would it take to change your mind about this stuff?
Why do I have to fucking do it? Why do
I The multi billion dollar companies do a dogshit job
of actually explaining this stuff or selling it. They lose
billions of dollars. But I'm the guy who has to
justify myself. Oh well, I'll keep doing it. Nevertheless, NES's
old At the bottom of this article's a far more
obvious payale horse. Open ai is planning to charge twenty
(07:59):
percent to thirty percent of pro customers the two hundred
dollars a month subscription that loses the money every time,
a higher price because of how many research queries they're
doing with Altmann, according to the information, suggesting some sort
of hey guessoir a la carte or pay as you
go approach. I want to be clear about something. This
is not a company that's cooking. This is not a
(08:21):
company that's worked out anything. Open Aiye is unprofitable, unsustainable,
and deeply, deeply lost. These are the actions of a
desperate company run by a desperate man. You have only
Sam Morman had a thoughtful friend to talk about all
these problems too,
Host
Ed Zitron
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com