# ChatGPT Work vs. Claude Cowork

An episode of [AI Chat: AI News on Claude, ChatGPT, Gemini, Grok, Perplexity](https://invisiblenetwork.ai/podcast/1669990482), 27 Aug 2026, 12 min.

Audio: https://content.rss.com/episodes/365073/3097681/ai-chat-podcast/2026_08_26_13_22_43_9f83e392-dca0-429b-bbba-8fd05bbcbff1.mp3

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! In this episode, we explore OpenAI's significant price reduction for ChatGPT Work and the implications for users. We also discuss the nuanced differences between ChatGPT and Claude, highlighting their unique strengths and weaknesses in real-world applications. Chapters 00:00 User Growth of ChatGPT Work 00:09 API Price Reduction Details 00:43 ChatGPT and Claude Comparisons 01:59 Workflow Experiences with Both 04:11 Long-Running Tasks Analysis 07:52 Conclusion and Future Tools Show Links • Get the top 80+ AI Models for $8.99 at AI Box: https://aibox.ai • Get the AI Chat Daily Newsletter: https://www.aichatdaily.com/newsletter

## Transcript

[0:02] Welcome to the AI Chat Podcast. I'm your host,
[0:04] Jaden Schaefer. Every day, I cover cutting -edge
[0:07] AI news and talk with the leaders behind it,
[0:09] breaking down what it means for your life and
[0:11] business. ChatGPT Work, the cloud co -op competitor,
[0:18] has reached 20 million users. This is according
[0:21] to their head of core products. OpenAI said that
[0:24] they are cutting the API prices 80 % with Luna,
[0:28] their Luna model, and they said that this is
[0:30] going to be a permanent... price correction.
[0:32] So if you're using Cloud, or if you're using
[0:34] ChatGPT work to get stuff done, and you switch
[0:36] the Luna model, you're going to use very little
[0:38] credits. And the reason why this is important
[0:41] is because for so long, you're only able to access,
[0:44] if you wanted to get access to ChatGPT work,
[0:47] you had to pay $100 and get on the enterprise
[0:49] tier $100 a month. But now ChatGPT work is going
[0:52] to be bundled with their $20 a month plus plan.
[0:54] So anyone basically is going to be able to get
[0:57] access to ChatGPT work. I think this is very
[0:59] smart because essentially at the end of the day,
[1:01] when it comes to the $20 plan, the $100 plan,
[1:04] the $200 plan, I mean, they're basically all
[1:06] the same thing. I'm sure there's some extra checkboxes
[1:08] that they say you're going to get, but the...
[1:10] product basically functions the exact same way.
[1:12] The difference is how much usage you get. And
[1:14] so I don't think there's any sense in saying,
[1:16] look, our, you know, our greatest feature is
[1:19] paywalled behind our more expensive tiers, because
[1:22] at the end of the day, they want their lower
[1:23] tier people to get access to this tool, which
[1:26] in my opinion, it's kind of this magical aha
[1:28] moment with AI. And then if you have a ton of
[1:31] projects, you're just going to upgrade to a higher
[1:33] tier to use them. And that's kind of where I
[1:35] found myself with Claude. I think Claude did
[1:37] that same strategy. And I was able to get access.
[1:40] I know a lot of people that use cloud co -work,
[1:42] the $20 a month, but at the end of the day, if
[1:44] you have a big project, you're just, you're going
[1:45] to be happy to pay a hundred or $200 a month
[1:47] to get it all done because the cost of hiring
[1:49] developers would have been thousands of dollars
[1:51] back in the day. So people just see a lot of
[1:53] the value there and not just developers, but
[1:56] there's so many different tasks you can do. There
[1:59] was an interesting quote I saw. So Sodio, who's
[2:02] their head of core product, said that the design
[2:05] philosophy is to get out of the way of the model,
[2:08] not put a whole bunch of like... AB choice questions
[2:11] in front of all the users and ask them to answer
[2:14] all of this, but really to have the model take
[2:15] a lead. I've actually found this. I actually
[2:18] use Claude cowork and chat GPT work both at the
[2:22] same time. I have both subscriptions for different
[2:25] reasons, which I could get into. But it was interesting
[2:27] because Ethan Mollick, the Wharton professor,
[2:30] who's, you know, big AI influencer, he was saying
[2:33] it's one of the biggest difference he sees between
[2:34] the two models is that Claude's cowork is going
[2:37] to give you a lot more decision points. It's
[2:39] going to ask. you a lot more questions. But that's
[2:42] not really the case with chat GPT. It's trying
[2:43] to get stuff done. And actually, I think there's
[2:45] pros and cons to that. Yesterday, I was working
[2:48] with both of the models at the same time on two
[2:50] different tabs. And I'm sure people are like,
[2:53] you know, people like to say everyone else's
[2:56] workflows are dumb for whatever reason. I'm sure
[2:58] people would criticize that workflow that I have,
[3:00] but I have a maxed out $200 a month chat GPT
[3:03] work plan. And then I have a hundred dollar anthropic
[3:05] plan. The reason I had to get the a hundred dollar
[3:07] anthropic plan is because I did notice there
[3:09] was like some bugs with the. chat GPT plan, I've
[3:12] got a few times where it's not it's not able
[3:15] to access my browser, and I have to restart my
[3:17] whole computer, and then restart it. And anyways,
[3:19] it works. There was also an issue where it wasn't
[3:21] able to connect to the meta MCP for doing my
[3:24] meta ads, which I really need to do as a big
[3:25] project. So I also got Claude for that and a
[3:28] couple other things. So you know, hopefully they
[3:31] fix that. I do think at the end of the day, I
[3:33] prefer the chat GPT work experience to Claude
[3:37] shocker because I know I've been like the biggest
[3:39] Claude advocate. for at least nine months now.
[3:43] So I do think they have better experience. But
[3:45] like I mentioned, they got a couple of bugs.
[3:46] So, you know, you got to take that with a grain
[3:48] of salt, hopefully to get those figured. One
[3:50] thing I did notice that was different when I
[3:51] was working between the two is I would tell,
[3:53] I would ask, I would actually, a couple of times
[3:55] I gave Claude and Chachi the exact same prompt
[3:57] in co -work and work to see what the difference
[3:59] was. And Claude would... get like the first chunk
[4:03] of what I said done, and then it would get to
[4:05] a breaking point. And it's like, okay, I did
[4:06] X, Y, Z, would you like me to continue with the
[4:08] next thing? That's not bad. But if you really
[4:10] want to be hands off, it can it can be annoying.
[4:12] Whereas chat GPT work, I would be like, hey,
[4:14] we're going to do this because my goal at the
[4:16] end of the day is to post this to this other
[4:18] place. So I would like, I kind of gave it the
[4:20] broad project, but I gave it one specific thing.
[4:23] And sometimes you go a little too far. It would
[4:25] be like, okay, awesome. Like I, you know, created
[4:27] the plan and I have it all and actually went
[4:29] updated or, you know, I was like, I've already
[4:30] started uploading all the files to the end platform.
[4:33] I'm like, whoa, whoa, whoa. Like I need to like
[4:34] check it and make sure it like looks good. And
[4:36] I guess my prompt, I should have been like, do
[4:38] this and then let me know. And then I'll check
[4:40] it. And then once it's approved, then you can
[4:42] go upload it. I'm just so used to cloud automatically
[4:44] doing that. Now, is that a bad thing? I don't
[4:47] necessarily think so. I think you just have to
[4:49] understand that with chat GPT, if you, give it
[4:52] the overall vision. It's going to try to execute
[4:54] the whole thing, which actually can be like,
[4:56] honestly, very nice because the bane of my existence
[4:59] is when I'm on Claude and I'm like, hey, Claude,
[5:01] can you, you know, go get this project done for
[5:03] me? And then, you know, I turn around, you know,
[5:06] I go to the store or something for a couple hours
[5:08] and come back. And then I realized that it was
[5:10] stuck. It didn't do anything because it was like,
[5:12] sure, I can do that project. Do you want me to
[5:14] do this, this or that? And then it just has this
[5:15] like waiting answer screen that was open for
[5:18] two hours. And I click it and it's like, OK,
[5:20] now I'm getting started. And I'm like, ah, like.
[5:21] I really wish I could have just done that over
[5:23] the last two hours while I wasn't there. So ChatGPT
[5:25] doesn't seem to have that problem. Another thing
[5:27] that I really do like about ChatGPT work is if
[5:30] I say, hey, look, like try to get this project
[5:33] done over, like don't stop till you get this
[5:36] whole project done. I've given it really big
[5:37] projects before. Recently, I needed it to write.
[5:40] I was building a directory. for a project that
[5:44] i was working on and i had is it's like a church
[5:47] finder directory as part of my learn of christ
[5:50] bible study app that i'm building a website so
[5:52] i was making this directory but there's 350 000
[5:55] churches in america and i was like i was like
[5:58] hey i want to make like a directory um so of
[6:01] every church so people can like you know put
[6:02] in whatever they're looking for and it can direct
[6:04] them to some because i wasn't finding a super
[6:06] complete directory And the issue is with 350
[6:10] ,000 churches, I need to write a description,
[6:12] like a little description for each church. Like,
[6:14] oh, this is a Pentecostal church that's known
[6:16] for X, Y, Z things located in this area. People
[6:19] like this about it. Right. Just like some sort
[6:21] of like short blurb for everyone to give people
[6:23] some context. And so I was going to go to their
[6:25] website and get some information to write that.
[6:28] Well, writing 300 ,000. um kind of like directory
[6:32] descriptions is going to take a very long time
[6:35] to do especially with all of the research and
[6:37] i'm kind of used to with claude i'll give it
[6:39] like a big project like that and claude will
[6:42] get started and then be like okay like i've written
[6:45] like a thousand of them this is a big project
[6:47] are you sure you want me to continue and i might
[6:48] continue and then it will you know okay i've
[6:50] written a thousand more so i gotta do that 350
[6:52] times over the course of you know, a few days
[6:56] or a couple weeks in order to get that whole
[6:57] project done. That's kind of the workflow. And
[6:59] sometimes I'm like, don't stop until you get
[7:01] this whole thing done, whatever. The thing that
[7:03] I love about ChatGPT work is I literally have
[7:06] that prompt running and it sets up like a long
[7:10] running task and it tracks it and it has a timer
[7:13] on there and I'll come back and it's like, we've
[7:16] been running this task for 24 hours and 14 minutes
[7:19] and we're on batch, you know, 5 ,000 of, you
[7:23] know, X amount of batches that we're doing for
[7:26] this. ChatGPT has the ability to do these like
[7:29] really long standing tasks. I know that's not
[7:31] like a model limitation. It's not like intrinsically
[7:33] ChatGPT's models are smarter than Claude's. It's
[7:35] just that Claude's harness hasn't really given
[7:38] Claude the ability to as easily get those long
[7:41] running tasks done that ChatGPT has done. So
[7:44] right, there's pros and cons of every model.
[7:46] That's one thing that I really appreciate it
[7:48] for if there's something that I need done. super,
[7:50] super well, I feel like the absolute ultra max
[7:53] top of the line AI model for open AI right now,
[7:56] I feel like it's a little bit better than codec
[7:58] or than Claude, you know, code, but I'm also
[8:02] not really using fable five that much because
[8:04] it just burns my credits. And I just I don't
[8:06] know, I kind of got burnt out of there. And so
[8:08] mostly, I'm just using mostly I'm just using
[8:11] the top models from chat GPT when I need to get
[8:13] some of those long standing things done. So pros
[8:15] and cons of each model at the end of the day,
[8:17] chat GPT work, in my opinion, right as of right
[8:20] now is a slightly better experience, but Claude
[8:23] work is definitely better at making those connectors
[8:27] and MCP connections. It's a little bit smoother
[8:29] of a process. And overall, I think they've had
[8:32] a little bit more time to refine the app to what
[8:34] people want. I think ChatGP is going to probably
[8:37] catch up soon and maybe, you know, Claude will
[8:39] catch up to some of the features. So they're
[8:41] pretty similar. And I guess it really depends
[8:43] which one you're more familiar with or which
[8:45] one you have more context with. But I don't think,
[8:46] you know, if you've been using ChatGP this whole
[8:48] time and someone's like, oh, you got to try Claude.
[8:50] At this point, I don't think you're missing out
[8:52] on anything. and vice versa. If you're really
[8:54] big on Claude and you don't want to switch to
[8:56] Chai Chippity, I don't think you're missing out.
[8:58] So I wouldn't recommend switching from whatever
[8:59] you have, but there are pros and cons and I am
[9:01] actively using both. And I will say the results
[9:04] are slightly different. So one thing that I do
[9:10] frequently for a lot of different projects that
[9:11] I run is I will tell it to do like an SEO audit
[9:16] of a lot of my sites. I have a couple of big
[9:17] sites that I'm growing through just like organic.
[9:21] seo kind of google search results one of those
[9:24] is aichatdaily .com it's my ai news site where
[9:28] i publish about 20 different articles a day and
[9:30] a lot of the content i talk about on here comes
[9:32] from that site um and so i always am like looking
[9:36] for ways to help that rank better in google specifically
[9:39] i'm trying to rank in the google news feed which
[9:41] it's really cool because you know when when i
[9:43] can get a story published faster than any of
[9:45] the major journals um because i'm like using
[9:48] ai to help assist me in a lot of that i will
[9:50] rank on Google very quickly. And I see like explosive
[9:53] growth to my website. So something I do that's
[9:56] really important for that is making sure my website
[9:58] is optimized and very like fast, like fast load
[10:00] times, fast to deliver the content. Everything's
[10:04] really high quality. And the SEO is like a lot
[10:06] of different optimization stuff I do. So I asked
[10:09] Claude and I asked ChatGPT both like, hey, do
[10:11] an SEO audit. I try to do this once a week for
[10:14] this project and for my Bible study app that
[10:17] I'm working on, Learner of Christ. And I'll tell
[10:18] like, say like, hey, go to my Google Search Console,
[10:21] download all of my reports there of all my analytics,
[10:25] and then give me like an overall SEO strategy.
[10:28] Make sure that you go and tell me like, give
[10:32] me recommendations of what new content to write,
[10:34] what new types of things to cover, and also what
[10:37] optimizations I should do to my page. And it's
[10:39] really interesting because I will actually get
[10:41] different results from from both of those two
[10:43] models where one will catch something the other
[10:45] didn't and i mean even the same model back to
[10:48] back will catch things that it like i'll do an
[10:50] audit and then i could literally turn around
[10:52] in a new chat thread on clod after getting clod
[10:55] to do an audit and say do another audit and it
[10:56] might catch another couple things so i think
[10:59] at the end of the day having both models is a
[11:01] bit of a bonus a bit of a superpower i've seen
[11:03] a lot of people in the early days say like Like,
[11:06] oh, I'm using ChatGPT to clean up all the bugs
[11:08] that Claude left behind. And like, yes, but also
[11:12] you could use Claude to clean up bugs that ChatGPT
[11:15] left behind. So it's not like one model's, you
[11:17] know, completely goaded and better than the other.
[11:19] I think using both is a superpower if you have
[11:22] it, but if not, it's not the end of the world.
[11:24] These really are incredibly powerful frontier
[11:26] models that are getting a lot of work done and
[11:29] doing... a ton. So both of them are fantastic
[11:32] at the end of the day. Hey, thanks for joining
[11:34] me for this breakdown of everything happening
[11:36] with Claude and ChatGPT. This is an exciting
[11:39] time. There's a ton of new tools coming out.
[11:41] I'm excited to test out some of the other competitors
[11:44] more heavily. I know that SpaceX has their new
[11:47] Grok bot that came out. And I know that Google
[11:49] is working on some great competitors in this
[11:51] field as well. So I'll keep you guys up to date
[11:53] as those come out. These are kind of the tools
[11:54] I love and I'm using every day. So I have a pretty
[11:57] great, pretty deep understanding. of some of
[12:00] the pros and cons, and I'll make sure to keep
[12:01] you guys updated on that. Thanks for tuning in.
[12:03] Make sure to leave the podcast a rating and review
[12:05] wherever you get your podcasts. And make sure
[12:08] to go check out AIbox .ai if you want to get
[12:11] the ability to generate images, audio, text,
[12:13] video, all inside of Cloud or all 80 of the most
[12:16] popular AI models in the world generating inside
[12:19] of ChatGPT or any other AI model. It's AIbox
[12:22] .ai slash MCP.

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.