Latest / In the Loop AI / THIS WEEK IN AI: Fable 5 Returns, OpenAI Might Nationalize, Meta Sells Compute

Transcript

THIS WEEK IN AI: Fable 5 Returns, OpenAI Might Nationalize, Meta Sells Compute · In the Loop AI

0:00
Ejaaz: The King is officially back. Anthropix flagship model Fable 5 is finally available
0:04
Ejaaz: for everyone after being banned for almost two weeks by the US government.
0:08
Ejaaz: It is still the world's best model. I've been using it in the 12 to 24 hours
0:12
Ejaaz: since it's been launched. I'm super excited about it.
0:15
Ejaaz: There's a bunch of demos that we want to get into almost immediately,
0:18
Ejaaz: starting off with my favorite kinship as a child, Hogwarts and Harry Potter.
0:24
Ejaaz: What you're seeing in front of you is a completely simulated fabrication of
0:28
Ejaaz: the entire Harry Potter universe starting off with Hogwarts and you'll notice
0:32
Ejaaz: that I can click into this universe immediately and start looking around.
0:37
Ejaaz: It's important to mention that this was built from a single prompt.
0:40
Ejaaz: I'm not entirely sure how long it took but Fable 5 did this in one complete
0:45
Ejaaz: shot and if I press B you'll notice something pretty spectacular.
0:50
Ejaaz: I'm on a broom and I'm flying around in probably my favorite childhood universe.
0:54
Ejaaz: Josh, I don't know if you're a Harry Potter fan, but this is pretty freaking awesome.
0:58
Josh: It's unbelievable. And I'm not sure if it was one shot. This is pretty impressive
1:01
Josh: for one shot, but it was absolutely generated with Fable 5.
1:05
Josh: And it's a testament to how impressive the model is and how happy I am that it's back.
1:09
Josh: Thank God. Holy shit. It's been a long couple of weeks without you.
1:12
Josh: I mean, what's most amazing about this, I find, is that not only is it able
1:17
Josh: to generate the map of Hogwarts, but it's able to generate the unique pieces.
1:21
Josh: Like you just went into the dining room. This is the weeping willow tree.
1:24
Josh: It has sound design built into it.
1:26
Josh: It understands the physics and the scaling of the way things should be built.
1:30
Josh: And look, and like, now you're going over to play Quidditch.
1:33
Josh: It's this like unbelievable awareness of the world and this unbelievable ability
1:38
Josh: to generate these visual elements. Like this is so amazing. The fact that you're
1:41
Josh: playing this in your browser
1:43
Josh: just by someone who prompted the Fable a couple of times and then deployed this.
1:46
Josh: And you could see it's like pretty accurate. You have Hogwarts,
1:48
Josh: you have the dining hall. Everyone has Fable until July 7th,
1:51
Josh: I believe the day is. 7th. In which the limits get...
1:54
Ejaaz: Unpaid subscriptions.
1:56
Josh: And then the limits change a little bit. And this is subject to change as always.
1:59
Josh: But currently right now, you have about a four to five day window to really
2:02
Josh: play around with this as freely as it's likely ever going to be.
2:05
Josh: The intention of showing this demo is to showcase that the possibilities of
2:09
Josh: what you can make with this model really are pretty much endless.
2:11
Josh: And I think the challenge to everyone now is now that there is this model that's
2:14
Josh: out that's truly unbelievable in capability relative to anything we've seen,
2:18
Josh: it's to go out and actually try unique and fun things.
2:20
Josh: Like I would have never thought to go off and make a Hogwarts game that I can go and play.
2:25
Josh: I mean, traditionally you buy this game. I have a Hogwarts game that I bought
2:28
Josh: for 75, $80 on Xbox to play around with. So what types of difficult challenges
2:33
Josh: do you have that you'd like to try and that you never thought were possible?
2:37
Josh: And like, those are the things that I would encourage trying out with Paypal 5.
2:39
Josh: Really amazing model very cool you can see here the second demo this is all
2:42
Josh: about physics and the physics understanding and it shows it on a relative basis
2:45
Josh: compared to the other models and it's just so far superior in its understanding
2:49
Josh: of the worlds of physics so
2:51
Josh: a lot of these are visual examples it does really well on basically all other types of work so
2:57
Josh: That's the thing is just try it out, see what you can do. I love this intro
3:00
Josh: post that they shared Fable is back and it's now showing and the whole sign
3:04
Josh: lights up and it's like, man, we are so back. So really exciting day.
3:07
Josh: If you were listening to this, go track Fable, go play around with it.
3:10
Josh: It is an incredible model.
3:12
Ejaaz: I want to talk about what has changed because obviously like we can't ignore
3:16
Ejaaz: the elephant in the room, which is this thing was banned for about two weeks.
3:19
Ejaaz: The US government specifically banned it because it presented itself as a potential
3:23
Ejaaz: cybersecurity threat across all national defense systems, which of course is very important.
3:27
Ejaaz: The TLDR of their blog post is essentially they have added further safeguards
3:32
Ejaaz: and more monitoring as to how people use the specific tool.
3:36
Ejaaz: They'll be looking at every single prompt in terms of whether it could potentially
3:40
Ejaaz: be an attack vector on the US or anyone in general.
3:43
Ejaaz: And they've been working closely with partners from Glasswing,
3:46
Ejaaz: their collective, which they set up once they created Mythos to make sure that
3:50
Ejaaz: their defense systems are hardened appropriately against this model.
3:54
Ejaaz: That being said, there was a lot of fear mongering around the re-release of
3:58
Ejaaz: this model. There was talks of potential KYC verification or potentially only
4:03
Ejaaz: allowing people in the US or American-born people to get access to this.
4:07
Ejaaz: I'm happy to say that all of that has been dismissed.
4:10
Ejaaz: The government and Anthropic, as well as OpenAI and a bunch of other Frontier
4:13
Ejaaz: AI labs, have been working pretty closely together to form some form of a framework
4:19
Ejaaz: for any future model release such that
4:22
Ejaaz: Everyone can be banned against it, but it's released in a more secure way that
4:27
Ejaaz: people can still benefit from the access of it and not be used in a malicious
4:30
Ejaaz: way. So I'm happy about this. This is Fable 5 back in all of its glory.
4:34
Ejaaz: I'm using it across all of my chats right now. Opus 4.8, you have been a pleasure
4:38
Ejaaz: to work with, but sorry, I need to work with your older brother at this point.
4:41
Ejaaz: But I'm excited about it. You have until July 7th, as Josh mentioned.
4:45
Ejaaz: So use it as much as you can.
4:46
Ejaaz: It switches to usage based pricing depending on what task you have.
4:51
Ejaaz: But it's great to have it back um josh what have you been using it for personally i'm
4:54
Josh: Curious i've been using it for all of my hardest challenges and all of the things
4:58
Josh: that i do mostly day to day a lot of um just like research a lot of production,
5:03
Josh: pretty much any challenge that i have i have switched over to fable and i noticed
5:08
Josh: that it does it much better,
5:09
Josh: so in the case that we're researching a topic for a show for example it will
5:12
Josh: go much deeper it's more comprehensive you can kind of read through its reasoning
5:16
Josh: and understand how it comes to conclusions and it's a really good thought partner
5:19
Josh: at least for me in a lot of cases,
5:21
Josh: a big problem that i have again is figuring out where the edges of this thing
5:25
Josh: are and what i actually can use it for to be a little bit more creative it's
5:28
Josh: like okay i shouldn't really be using this as an extension of opus what new use cases can i try
5:34
Josh: and i think that's what i'm going to be spending a lot of time this weekend trying to figure out is
5:37
Josh: what are the harry potter like demos that i could use that i'm actually really
5:40
Josh: interested in building myself.
5:42
Ejaaz: Yeah so one thing i've noticed is um i've never had a personal website before.
5:46
Ejaaz: And I've thought like, hey, why don't I just give it a go and try and figure it out?
5:50
Ejaaz: Now, the one bit of critique that I've always had with any AI model up until Fable 5 is,
5:56
Ejaaz: the taste is just terrible. I'm sure you've tried this before,
5:59
Ejaaz: but you're like, hey, like, build me a website that looks something like the
6:02
Ejaaz: Airbnb website, you can share screenshot demos.
6:04
Ejaaz: And it just kind of does a terrible job or a janky job.
6:07
Ejaaz: Fable 5 is the first model that I've worked with that
6:10
Ejaaz: has this visual intelligence, which is super weird, because I can't currently
6:15
Ejaaz: use Claude right now to generate me an image, I can use ChatGPT to do that,
6:18
Ejaaz: I can use Google Gemini to do that, Nano Banana, but I can't do that with
6:23
Ejaaz: Claude specifically right now as an LLM.
6:25
Ejaaz: But yet it understands what I see so well.
6:29
Ejaaz: So what I've realized is if I'm using it to code up some form of a visual artifact,
6:33
Ejaaz: and we use visual artifacts a lot on our show as props or materials,
6:37
Ejaaz: it does an amazing job and it understands taste.
6:39
Ejaaz: It doesn't look like generic AI slop anymore. So I definitely recommend folks
6:43
Ejaaz: go and check that out. Now, one thing that we have to mention on the show that we noticed pop up was
6:49
Ejaaz: There was a discussion once Fable 5 was re-released because people started coding
6:53
Ejaaz: with this thing and they updated their recent version.
6:55
Ejaaz: And they noticed that there was some form of like, they described it as a spyware-like
7:00
Ejaaz: code in CloudCode that would basically track certain users' kind of activity on what they were using.
7:06
Ejaaz: And obviously there's a big reason for this because a lot of foreign state-based
7:12
Ejaaz: labs, such as in China specifically, were distilling Cloud models to build their
7:16
Ejaaz: own models, which is strictly against Anthropic policy.
7:19
Ejaaz: And what we got was a very quick dismissal from Darek himself at Anthropic saying
7:24
Ejaaz: that this was an experiment that we were launched to basically prevent distillation
7:29
Ejaaz: attacks from foreign adversaries.
7:30
Ejaaz: Now, if you're Anthropic and if you've been following the news,
7:32
Ejaaz: especially on Limitless, you'll know that a bunch of the Chinese open source
7:37
Ejaaz: models have effectively attained their status because they've distilled frontier
7:41
Ejaaz: models from American labs such as OpenAI and Anthropic.
7:44
Ejaaz: There was recent reports, I think this week of Alibaba's Quen doing the exact
7:48
Ejaaz: same thing. There was an official announcement from Anthropic.
7:50
Ejaaz: So it's good to see us taking these steps and measures to prevent such attacks.
7:54
Ejaaz: This is part of the whole safeguard approach, which is new with Fable 5.
7:58
Josh: Yes, big week for Anthropic. In addition to that, they also released Sonnet
8:01
Josh: 5 and Cloud Science, two major releases.
8:05
Josh: Sonnet 5, as everyone knows, is the kind of smaller model. It's the follow-up to Sonnet 4.6.
8:09
Josh: In fact, it's priced a little bit lower than 4.6, at least for the intermediary up until August 31st.
8:15
Josh: And it really is just like the entry-level model and the model you'll find that
8:19
Josh: Fable probably best wraps to.
8:21
Josh: So when you're using Fable, sometimes it gets expensive. You're using it for
8:24
Josh: very long-running tasks. Sometimes not everything needs frontier intelligence.
8:28
Josh: Sonnet 5 is a pretty good substitute for it to defer and start to offload tokens and requests to.
8:32
Josh: And it seems like it's just like, you know, a solid upgrade.
8:35
Josh: There's nothing exceptionally incredible about it, but it is there.
8:39
Josh: It is cheaply priced and it is very strong in the parts that it needs to be.
8:44
Ejaaz: I actually really like the look of Sonnet 5 for one particular reason.
8:48
Ejaaz: Over the last week, people have been praising GLM 5.2, which is a Chinese model from the lab Jipu.
8:55
Ejaaz: Way and we did an episode on this and it's a very impressive model um but
9:00
Ejaaz: but my take on this is i think american frontier labs aren't just going to sit
9:04
Ejaaz: around and let chinese ai labs outcompete them on cost specifically
9:08
Ejaaz: so they have the best cards to play which is if you have the frontier ai model
9:12
Ejaaz: such as claude fable 5 or chat gpt 5.6 sol
9:16
Ejaaz: you can just distill your flagship model into a much cheaper model and that's
9:19
Ejaaz: what we're seeing with sonnet 5 it does opus 4.8's capability but at a fraction of the cost.
9:25
Ejaaz: And so I think we're going to see this trend continue until we find that perfect
9:28
Ejaaz: medium in between how cheap GLM 5.2 is and how expensive Fable 5 is,
9:33
Ejaaz: and people will just stay within the Claude ecosystem.
9:35
Ejaaz: And I think it's a smart approach, and I'm happy to see it. But you mentioned
9:39
Ejaaz: Claude Science as well, Josh.
9:40
Josh: Super cool, yeah.
9:41
Ejaaz: As a former bio nerd that did a four-year biology degree and spent a lot of
9:47
Ejaaz: time studying genetics, I wished I had this tool.
9:51
Ejaaz: This is basically the co-pilot for any kind of scientific research that you
9:56
Ejaaz: do. It's a model that basically is fine-tuned and built to evaluate scientific
10:01
Ejaaz: primary research, help you do all your data analysis.
10:04
Ejaaz: And it's been used across a ton of different things. I mean,
10:07
Ejaaz: if you look at this guy called Andre, who has been playing around with cloud science.
10:12
Ejaaz: He installed it with Ligand AI, which is basically an MCP tool,
10:18
Ejaaz: which basically allows him to analyze different molecules.
10:21
Ejaaz: One of the big things when you're kind of figuring out whether an antibody or
10:25
Ejaaz: a molecule is good enough to fight a disease is you need to know whether it
10:29
Ejaaz: can attack a specific active site, which is like basically the vulnerable site of a
10:35
Ejaaz: bacterial virus or whatever that might be.
10:37
Ejaaz: And this system with cloud science
10:39
Ejaaz: basically helps you identify it um de novo which is like super cool
10:43
Josh: Yeah i think this is probably the most interesting and exciting release of the
10:47
Josh: week granted fable 5 already released so we're just doing it again um but cloud
10:51
Josh: science you could think of it kind of like.
10:53
Josh: The way that cloud code is intended to support engineering cloud science is intended
10:58
Josh: to support science and that's all of what this is about it integrates a very
11:01
Josh: specific set of tools it allows you to audit the outputs and anthropic is actually
11:06
Josh: using it they announced that they're using it to pursue their own drug discovery
11:08
Josh: programs for some rare neglected diseases. And people have been using it.
11:12
Josh: The Allen Institute now completes 100 plus page literature reviews that would
11:16
Josh: previously take up to two years.
11:18
Josh: And then some analysis now run in a 10th of the time of what they previously
11:22
Josh: did, thanks to this new software harness that they've introduced for science based purposes.
11:27
Josh: So when you combine a model like Fable into a harness like Claude Science,
11:31
Josh: you get some pretty amazing technological.
11:34
Josh: Innovation or at least the opportunity to do so if you are feeling so inclined
11:38
Josh: to through people who just like are involved in research in general so i think
11:41
Josh: this is super exciting for the science world i think we're going to see a lot
11:44
Josh: of really interesting things coming from this and that moves us on to our next
11:47
Josh: topic which is science related from
11:49
Josh: meta believe it or not the uh the metaverse company that was supposed to make
11:53
Josh: metaverse they were all supposed to make ai well it turns out the thing they're
11:56
Josh: releasing this week is is biological based and it is a brain-to-text decoder.
12:01
Josh: What is a brain-to-text decoder? Well, basically, topologically,
12:04
Josh: they insert some, or they put some electrodes on your head, and correct me if
12:07
Josh: I'm wrong, you guys, but they have actually managed to hit 61% word accuracy
12:12
Josh: versus 8% for previous non-invasive methods.
12:14
Josh: So, they're basically able to go into your brain, detect which neurons are firing,
12:20
Josh: and then derive a series of words that you are meaning to say without actually
12:24
Josh: saying those words. It's essentially reading your mind. Is that right?
12:26
Ejaaz: The best part about this is they only did this with nine volunteers,
12:30
Ejaaz: nine people, no, not even double digits.
12:32
Ejaaz: And they trained this model on basically 22,000 sentences. So comparably,
12:38
Ejaaz: if you look at like an LLM, which is honestly the frontier LLMs like Fable 5
12:42
Ejaaz: are upwards of like 15 trillion parameters, and they use a ton of compute.
12:45
Ejaaz: This is a small model in comparison, but it's equally so, so powerful.
12:50
Ejaaz: They basically are able to interpret different brain signals and convert that into text or words.
12:56
Ejaaz: The reason why this is so important is, well, firstly, there's the medical side
12:59
Ejaaz: of things. We've spoken about Neuralink quite a bit.
13:02
Ejaaz: This is kind of Meta's answer to that, at least in the initial types of research that they're doing.
13:07
Ejaaz: Now, application wise, I am curious what Meta is going to use this model for.
13:12
Ejaaz: Obviously, there's the medical side of things, but we know and we've spoken about this previously.
13:17
Ejaaz: They are working on a series of different brain models, as well as ECG-related
13:21
Ejaaz: devices that you can wear on your wrist and on your heart that can basically
13:25
Ejaaz: interpret everything that you're thinking and figure out what the best type
13:28
Ejaaz: of content is to produce for you.
13:30
Ejaaz: So it's an interesting thing where Facebook is a social media site.
13:34
Ejaaz: And one thing that they've spent a lot of time figuring out is how to get all
13:38
Ejaaz: the wonderful imaginative thoughts in your head out into the open world.
13:42
Ejaaz: They've done that with Instagram through you picking up a phone and taking pictures of stuff.
13:45
Ejaaz: They've done that with Facebook, translating your thoughts into cool ideas that
13:49
Ejaaz: you can share with the world.
13:50
Ejaaz: This seems to be a step in a direction where for those folks who aren't really
13:54
Ejaaz: good at, you know, using those skills to translate your thoughts into some kind
13:58
Ejaaz: of a vivid pitch or whatever that might be, you can now use this potential tool
14:01
Ejaaz: in the future for something like that.
14:02
Ejaaz: And listen, I have a more optimistic take for all these kinds of tools,
14:05
Ejaaz: which is, you know, they're going to use it for the advancement of humanity and good.
14:08
Ejaaz: But I don't know what meta specific purpose for that might be.
14:12
Ejaaz: But it's interesting to see nonetheless. And I like that this kind of research
14:15
Ejaaz: is being focused on despite LLM is being focused on being chatbots or generating
14:19
Ejaaz: images this is like advancing real science and you know between OpenAI's gene
14:23
Ejaaz: benchmark between Claude Science and between this this is great to see
14:26
Ejaaz: uh it's a great week for science in AI for sure
14:28
Josh: Yeah it's been really exciting it's cool to see progress and then I'm also just
14:31
Josh: like very confused at what Meta is doing because we're talking about science
14:34
Josh: now but then there was another big pivot that just happened where Meta is shifting
14:37
Josh: to cloud infra they're they're like pivoting to the SpaceX plan so it seems like
14:42
Josh: Meta over the last decade has just been this series of unfortunate pivots and
14:45
Josh: every time we pivot somewhere new and we try this new hard thing it doesn't
14:49
Josh: quite work out and they pivot again and it feels like a startup kind of trying to find some traction
14:53
Josh: this is the newest iteration of that in addition to the science breakthrough
14:56
Josh: that they're sharing in terms of like the topical neuron receiver stuff they're
14:59
Josh: also just going to start selling compute
15:01
Josh: so i don't know if this is bullish or bearish for their internal ai
15:05
Josh: program i have to imagine it's not that exciting because if it was going so
15:08
Josh: well they wouldn't have extra compute sell
15:10
Josh: but the idea here is at least that they will begin to sell their compute to
15:14
Josh: third-party companies similar to what spacex has been doing with their deals with anthropic and
15:18
Josh: who else did they sign a deal with google and i think it's just another instance
15:21
Josh: of them trying things throwing stuff against the wall and seeing what sticks
15:24
Josh: i don't know you just do you have any takes when you read this news.
15:27
Ejaaz: Yeah, I have an alternative take, which is, I think it's a smart move for Zuck.
15:32
Ejaaz: It is... Market loved it. Yeah, it's a potential massive revenue earner.
15:36
Ejaaz: The stock went up, I believe, like 6% to 8% on the day.
15:40
Ejaaz: And the reason is pretty simple. Meta has aggregated one of the biggest fleets
15:46
Ejaaz: of the most important metals needed for AI right now. And that is the GPUs.
15:51
Ejaaz: We said the same thing about SpaceX. So SpaceX has Colossus 1, 2, and 3 data centers.
15:58
Ejaaz: And that is roughly over, I think it's a million of NVIDIA's top GPUs.
16:04
Ejaaz: And so that fleet can be very attractive for competing labs,
16:07
Ejaaz: such as Anthropic and Cursor, who they are selling their GPUs to.
16:11
Ejaaz: Meta saw this and thought, hmm, that did pretty well for the stock price.
16:14
Ejaaz: Let me do the same over here and bring some money in to justify additional AI
16:19
Ejaaz: capex spend for future years. So I think this is a strategic move from Zuck to do this.
16:23
Ejaaz: And I have another bit of evidence which proves that this isn't necessarily
16:26
Ejaaz: Meta running for the hills, which is Meta signed a compute deal with Google.
16:32
Ejaaz: And Google this week announced that they have to restrict Meta's use of their
16:37
Ejaaz: Google Cloud because they're using too much of it.
16:39
Ejaaz: So my suspicion, and we'll find out in the Q2 earnings report from Meta in a
16:43
Ejaaz: few weeks' time, is Meta's allowing all their old GPUs to be sold for inference.
16:49
Ejaaz: And they're going to make a ton of money in the same way that SpaceX has charged
16:53
Ejaaz: a huge premium to Anthropic and Cursor for selling inference through their GPUs.
16:58
Ejaaz: Meta's going to do the same, make a bunch of money from that,
16:59
Ejaaz: and save all the Vera Rubins and the newer GPUs to train future Meta models.
17:03
Ejaaz: And we reported, I think, two weeks ago that Meta is working on a potential
17:07
Ejaaz: Mythos level model that's going to be available nine months from now.
17:10
Ejaaz: I know that's kind of late, but I still think that in this race,
17:13
Ejaaz: they're just very, very, very far behind.
17:16
Josh: Well, hey, all the power to them. I hope they figure it out.
17:18
Josh: Now we should probably pivot over to someone who has figured it out mostly, which is OpenAI.
17:22
Josh: And we've recorded an entire episode on this. They released 5.6 this week.
17:26
Josh: If you haven't listened, I would highly advise going to check out that episode you recorded.
17:29
Josh: The update now is that it's still not live. And by the time of reading this,
17:33
Josh: that might have changed. By the time you're listening to this,
17:35
Josh: hopefully that's changed. I think everyone's really excited to get their hands on it.
17:37
Josh: But they have not met the same fate that Fable has just yet.
17:41
Josh: So they have Sol, Terra, and Luna, all associated with GPT 5.6. they are
17:46
Josh: probably being used internally you have to imagine they are being used in an
17:49
Josh: internal research project
17:50
Josh: that is with select companies but they are not broadly available so we're still
17:54
Josh: at a standstill in that point but the real news that we got this week was this
17:57
Josh: five percent number that's been floating around and none of this is confirmed
18:00
Josh: by any means but there's a lot of reporting going on
18:02
Josh: in that open ai is proposing to hand the,
18:06
Josh: united states government five percent of the company just for free right they're
18:09
Josh: just gonna give it away yeah.
18:11
Ejaaz: Yeah um i was going to bed last night, and I thought we had our agenda all prepped
18:15
Ejaaz: and ready, Josh, for today's roundup.
18:17
Ejaaz: And then I saw this breaking news from Financial Times, which basically says
18:21
Ejaaz: that Sam Altman is reportedly shopping 5% of OpenAI to the US government to take ownership.
18:27
Ejaaz: And the craziest part about this is, I don't think he's even requesting an investment.
18:31
Ejaaz: He's just kind of like handing this over for the good of the world.
18:34
Ejaaz: And if you want to know some context as to why he might be doing this,
18:37
Ejaaz: I think there's two perspectives of this. Perspective number one is
18:42
Ejaaz: Sam believes deeply that governance and national government should be involved
18:48
Ejaaz: in how these AI models are shaped.
18:50
Ejaaz: And we know this because in an earlier blog post from Sam, I think about a year
18:54
Ejaaz: and a half ago, he said this. He said it might actually make a lot of sense
18:57
Ejaaz: for the government to take stakes in major AI labs.
19:00
Ejaaz: The reason for this is, as we've seen with Claude Mythos and Fable 5 being banned,
19:04
Ejaaz: it is incredibly important that the government gets access to these models pre-release
19:09
Ejaaz: so that they make sure that it's not a cataclysmic threat to government security systems and whatnot.
19:14
Ejaaz: So it seems to be a good perspective from that side of things.
19:18
Ejaaz: But on the other side of things, this could be Sam using a chess piece to basically
19:22
Ejaaz: get a political advantage when it comes to writing policy frameworks.
19:26
Ejaaz: As we know, Sam's been heavily involved in lobbying certain governments and
19:30
Ejaaz: laws in terms of making frameworks more favorable for OpenAI's types of models.
19:34
Ejaaz: We see other AI labs doing the similar types of things. If you were a critical
19:38
Ejaaz: person of Sam, you might think that this is the move that he's making.
19:42
Ejaaz: It's kind of weird. I don't know how I feel about the whole nationalization
19:46
Ejaaz: of the AI movement, but it certainly seems to be the trend that we're moving towards currently.
19:51
Josh: Well, I like the fact that it was at least voluntary. This was presumably not
19:55
Josh: forced. They're offering this up.
19:57
Josh: You have to assume, I mean, again, these are all reports. Nobody knows what's
20:00
Josh: actually going on, but the idea is that they would just make a generous donation
20:04
Josh: to the united states government and this makes sense when you look at it in
20:06
Josh: terms of alignment it's like what is the
20:09
Josh: most pressing issue that a lot of people are very upset about right now it's
20:12
Josh: ai and the downstream effects that they think it's going to cause on their life
20:16
Josh: they're scared they don't understand um they have no vested interest in this
20:20
Josh: and i think giving everyone a
20:22
Josh: even if it's a small amount of vested interest in the success of these companies is probably net good
20:27
Josh: If it could even change the sentiment a little bit, if it can get people aligned,
20:30
Josh: if they could feel good about these companies winning, because it means not
20:33
Josh: only are they winning, but the country is winning and everyone is kind of aligned
20:36
Josh: directionally on the success of these companies.
20:37
Josh: I think it creates a really compelling case for handing some over to the government
20:41
Josh: and giving it back to the people that are allowing you to do this.
20:45
Josh: They might just be feeling exceptionally confident this week because we have
20:48
Josh: another report, another rumor, again, unconfirmed.
20:50
Josh: All of this is unconfirmed. We are waving our hands around. We are putting our tinfoil hats on.
20:54
Josh: But allegedly there has been some sort of optimization achieved internally that
20:59
Josh: allows them to cut inference costs in half for models that was applied to that,
21:05
Josh: that's crazy because half of the costs on hundreds of billions to trillions
21:11
Josh: of dollars is not marginal that is a pretty serious innovation so what what
21:16
Josh: is going on here you just is this even possible.
21:19
Ejaaz: Yeah. So if this report is true, the analogy looks like the following.
21:24
Ejaaz: Right now, if you're a free user, ChatGPT, there's roughly around,
21:30
Ejaaz: I'm going to say 100,000 to 200,000 GPUs that are out there that are processing
21:35
Ejaaz: your prompts, your inference, your requests, giving you answers.
21:38
Ejaaz: Those GPUs, you know, I hate to mention it, are super, super expensive.
21:42
Ejaaz: Now, if this report is true, it's cut down that inference cost that cost to
21:48
Ejaaz: 50 percent which means that this is now a couple
21:51
Ejaaz: hundred of gpus and this is not a number that i'm making up this is a number
21:56
Ejaaz: that's in the informations report
21:58
Ejaaz: specifically a couple hundred gpus which means that if you can cut down costs
22:03
Ejaaz: that massively you now have free capital to spend on compute in other ways in
22:08
Ejaaz: open ai's case it's probably to train more models
22:10
Ejaaz: or to use inference for other types of usage such as inference at test time
22:14
Ejaaz: scaling, or whatever that might be.
22:15
Ejaaz: So this is a major jump for a few different reasons.
22:19
Ejaaz: Inference roughly makes up, I think, 60 to 70% of revenue earnings across top
22:25
Ejaaz: frontier labs right now. So
22:26
Ejaaz: if you're a frontier lab, a lot of your cost typically goes into training.
22:30
Ejaaz: A lot of the money that you make comes in from subscriptions.
22:32
Ejaaz: Now it comes in from inference specifically for people using it to service agents
22:37
Ejaaz: that they're running on Autoloop for hours and hours on end.
22:40
Ejaaz: Agents have become basically the new trend. Enterprises are using it 24-7 at this point.
22:45
Ejaaz: So it's become its own type of economy. And inference is basically the bedrock
22:49
Ejaaz: that underpins all of that. Anthropic this quarter is reportedly meant to be
22:54
Ejaaz: the first AI lab that goes profitable in Q2.
22:57
Ejaaz: A large part of this is achieved through the cost or the revenue margins that
23:01
Ejaaz: they earn off of inference specifically. It's roughly around something crazy
23:04
Ejaaz: like 80%. I remember Christian Rao mentioned it on a podcast previously.
23:07
Ejaaz: So the point is inference is very important. And any way that you can cut costs
23:12
Ejaaz: down on that and extend your profit margin is a huge deal.
23:16
Ejaaz: And if OpenAI has found some kind of a breakthrough, and I don't know how they've
23:19
Ejaaz: done this. I'm looking forward to actually an official announcement.
23:21
Ejaaz: This would be a huge leap for them in terms of getting more capital to acquire
23:25
Ejaaz: more compute to build better models.
23:27
Josh: We do have a confirmed breakthrough this week. There is actually something that
23:30
Josh: has been publicly announced and it comes in the form of this company that just
23:34
Josh: exited stealth named Etched.
23:36
Josh: And Etched, I saw the post and I was like, oh, all right, cool.
23:39
Josh: It's like another one of those things. It's like they think they're going to
23:42
Josh: change the world and they're going to solve the GPU problem and tokens and price
23:46
Josh: per watt and they're going to solve all the efficiency gains.
23:48
Josh: And I was like, absolutely not. These guys are like in their early to mid 20s.
23:52
Josh: They have no experience doing this.
23:54
Josh: Surely this is just another one of those things that is a headline and then
23:58
Josh: it devolves into nothing but i started to read about what they were doing and
24:01
Josh: i started to listen to a few of the podcast episodes that these founders were on.
24:04
Josh: And i was shocked at how competent and incredible the product is and how talented
24:11
Josh: the density of this team is and everything about this
24:15
Josh: story turned into this like unbelievable breakthrough i was like wait holy shit
24:19
Josh: this is this is like pretty amazing and pretty magical and i think so much so
24:22
Josh: that we're probably going to record an entire episode about this company and
24:25
Josh: this general state of inference as it really sees breakthroughs probably next week.
24:29
Josh: But just a TLDR on Etch is they've basically figured out how to create...
24:33
Josh: A new chipset from the ground up first principles vertically integrated meaning
24:36
Josh: they're building all these things from scratch they're integrating them all
24:39
Josh: into a single unit their product is the way in which they manufacture
24:43
Josh: this server rack and the server rack is the delivered good but within that requires
24:47
Josh: cooling and chip architecture and all of the plumbing as it connects to everything
24:51
Josh: and they've created this series of breakthroughs using
24:53
Josh: low voltage inference which is one so they're able to squeeze out a lot more
24:56
Josh: performance per watt generally on a gpu like a blackwell chip you are getting
25:01
Josh: maybe 50% of its theoretical peak because if you go to 100%, the chip melts.
25:05
Josh: So being able to squeeze more tokens per watt is a really impressive thing.
25:09
Josh: They figured that out. They figured out a memory problem where normally you're
25:12
Josh: constrained to the memory on a chip.
25:14
Josh: Well, they've constrained it to the memory of the cluster. So the cluster becomes
25:17
Josh: one single coherent piece of memory.
25:19
Josh: And there's a series of these breakthroughs that they've kind of discovered
25:22
Josh: that were really impressive.
25:24
Josh: And then you look at the team and you're like, oh, wow, everyone on this team
25:27
Josh: is like shockingly competent and have been working in the industry for many,
25:30
Josh: many decades. In fact, I think the youngest people are the founders.
25:33
Josh: So Etch seems like a grand slam.
25:35
Josh: Stay tuned. We're going to have a lot more to say about this company.
25:38
Josh: It was shocking. It was really cool to see this week.
25:40
Ejaaz: Well, just to kind of like sing their praises a bit more, these guys have raised,
25:44
Ejaaz: I think, $800 million, but they already have a billion dollars worth of orders
25:49
Ejaaz: for a chip that currently doesn't
25:51
Ejaaz: exist yet and is currently going to go live by the end of the year.
25:54
Ejaaz: The backers for this team are pretty insane. You've got Peter Thiel,
25:58
Ejaaz: you've got Jane Street, but then you've got TSMC themselves.
26:01
Ejaaz: As far as I'm concerned, TSMC is the company that is required for NVIDIA to
26:05
Ejaaz: stay alive at the moment, they do all the manufacturing processes around all their GPUs.
26:09
Ejaaz: And TSMC taking an equity stake in a competitor to NVIDIA in the first instance,
26:15
Ejaaz: as far as I'm aware, is a huge deal.
26:17
Ejaaz: Now, you mentioned what the breakthrough is specifically, I think,
26:21
Ejaaz: When we talk about other companies that have made inference specific chips,
26:24
Ejaaz: because I'm sure you're thinking about this, listening to this,
26:26
Ejaaz: we've spoken about Cerebrus that went public in a huge IPO about a month ago,
26:30
Ejaaz: and they created an inference check.
26:32
Ejaaz: And it was the size of like a couple of plates or something crazy like that.
26:36
Ejaaz: Their breakthrough was effectively designing a chip that would allow your AI
26:39
Ejaaz: models to work incredibly faster, but it would use a lot of power.
26:43
Ejaaz: It was quite expensive to use.
26:45
Ejaaz: The difference with this chip that Etch supposedly is going to create and release
26:49
Ejaaz: at the end of this year is it uses low voltage, which means that it uses 75%
26:54
Ejaaz: less power for the same amount of inference output. That's the breakthrough that they've made.
26:59
Ejaaz: And if this company can pull it off. And like you said, Josh,
27:01
Ejaaz: it's like a bunch of Harvard dropouts.
27:03
Ejaaz: This would be an incredibly competitive threat to NVIDIA. Because as far as
27:07
Ejaaz: I'm concerned, NVIDIA is focused on general purpose GPUs, which are used for
27:11
Ejaaz: training specifically.
27:12
Ejaaz: But the inference market is pretty open to anyone and everyone.
27:15
Ejaaz: That's why you see Cerebrus, Grok, and now Etched come to light.
27:18
Ejaaz: So it's pretty cool to see.
27:19
Josh: It's so exciting. It's rare that a company comes around that's actually new
27:22
Josh: and novel and making real breakthroughs. So I'm stoked about this company.
27:25
Josh: I could not be more excited. I would love to participate. I wish they had the
27:28
Josh: ability to buy some shares because my God, what a monster of a company that's going to be.
27:33
Josh: For the people who do own shares in public markets, chances are they might not
27:36
Josh: have done too hot this week.
27:38
Josh: The memory stocks have been getting hit. People haven't really liked this.
27:43
Josh: And we have this post on screen that's at 2 million views.
27:46
Josh: And there's no way that this post moved memory markets, right?
27:50
Ejaaz: I think it did. That's insane. I was racking my head to try and figure out why
27:56
Ejaaz: memory stocks were down 10% yesterday.
27:58
Ejaaz: It was one of the most aggressive dumps we've seen of recent,
28:01
Ejaaz: especially after a meteoric rise in memory stocks over the last year,
28:05
Ejaaz: pretty much is up on average between three to 500%, depending on which memory provider you pick.
28:10
Ejaaz: Andrew Curran is one of my favorite accounts on X. If you don't follow him,
28:13
Ejaaz: definitely give him a follow.
28:14
Ejaaz: He posts this very cryptic tweet. He goes, I'm posting this prediction now so I can quote it later.
28:19
Ejaaz: There has been a significant breakthrough in architecture, specifically around
28:22
Ejaaz: memory efficiency, not by one of the big labs, but a team that spun out of OpenAI.
28:27
Ejaaz: This will probably be announced soon.
28:29
Ejaaz: And this led on everyone on a spiral in social media.
28:34
Ejaaz: Now, I have a bit of tea because I have some sources which reported a few things to me.
28:39
Ejaaz: Now, apparently the company that he was referencing is this company called Core
28:43
Ejaaz: Automation. What this breakthrough is,
28:46
Ejaaz: He has no idea. We have no idea. So right now, it's just word of mouth.
28:50
Ejaaz: And if core automation wants to make an announcement, they can.
28:53
Ejaaz: But if this announcement is true, it would mean that AI models largely require
28:58
Ejaaz: much less memory. Now, why is that a big deal?
29:01
Ejaaz: The most expensive material that is required to build GPUs and train models
29:05
Ejaaz: and inference models right now is memory. It roughly makes up around 50 to 60% of the cost.
29:10
Ejaaz: And the price per unit of memory has skyrocketed over the last year specifically.
29:16
Ejaaz: Margins on Micron's recent earnings report, for example, has hit 80%.
29:21
Ejaaz: Traditionally, this is around 30%.
29:23
Ejaaz: So the fact that it is over double this is absolutely insane.
29:26
Ejaaz: And the margins keep going up, the prices keep going up. In fact,
29:29
Ejaaz: Apple announced recently that they are lobbying the Trump administration to
29:33
Ejaaz: buy memory from Chinese companies, which is a big no-no in terms of import-export
29:38
Ejaaz: controls that Trump and previous US governments have placed.
29:41
Ejaaz: We don't know if this is going to get passed, but the point is companies like
29:45
Ejaaz: Apple need to now increase the price of their products because they can't get
29:49
Ejaaz: their hands on enough memory.
29:50
Ejaaz: So they need to end up paying more, which means they need to charge you as a
29:53
Ejaaz: consumer more. So memory is in very scarce supply.
29:56
Ejaaz: My take on all of this is it's being blown out of proportion.
29:59
Ejaaz: Memory is an incredibly constrained commodity, and there's no way to scale it.
30:03
Ejaaz: You need a bunch of very complex machinery and chip fabs to be able to do this
30:06
Ejaaz: and we are nowhere near i don't see the supply scarcity being alleviated until
30:10
Ejaaz: at least 2028 by the end of that year
30:14
Josh: Well, yeah, we know this. The markets are very short-sighted always.
30:16
Josh: When there's headline news, they're going to overreact. There's going to be
30:19
Josh: this knee-jerk reaction. That's what we've seen. In fact, there isn't even really
30:22
Josh: headline news. This is just kind of all speculation.
30:24
Josh: It's funny, I'm looking at Core Automation and their URL is coreauto.com.
30:29
Josh: It looks like a freaking car manufacturing company. but it's like okay sure
30:33
Josh: like they have these breakthroughs but how even in the case that they do have
30:36
Josh: a breakthrough open ai or anyone what is the
30:38
Josh: the lag time between breakthrough to actual implementation where it is felt
30:42
Josh: in the market probably fairly long there's probably
30:45
Josh: i mean certainly a tremendous amount of demand for memory in everything that's
30:48
Josh: being built not only on the software side but also on the hardware side if you're
30:51
Josh: building robots if you're building automated hardware manufacturing,
30:54
Josh: all of this requires requires a tremendous amount of memory so is Is there a memory bubble popping?
31:00
Josh: Like, probably not. But again, not financial advice.
31:03
Josh: Who the hell knows? Talking about robots, though, we do have a robot topic because,
31:07
Josh: we're getting some consumer robots like robots are becoming a thing.
31:10
Ejaaz: Look at this cutie pie, Josh. You want this in your apartment?
31:13
Ejaaz: You want this in your home?
31:14
Josh: It's pretty adorable. Like, do I want that in my apartment? Honestly,
31:16
Josh: no. My apartment isn't big enough for the two of us.
31:19
Josh: I feel like we'd probably bump into each other and it probably wouldn't be.
31:22
Ejaaz: This thing is massive.
31:24
Josh: It's massive. And, you know, it's still moving a little slow.
31:27
Josh: It's like not really doing everything that i need for my liking maybe if you
31:30
Josh: live in a big house it's fun to have this guy running around but i do think
31:33
Josh: it's cool and exciting that more companies are trying this in fact this one's
31:36
Josh: actually dare i say kind of cute
31:38
Josh: it looks like this fun little robot that you can have walking around the house
31:40
Josh: you can get it in different colors and you can pre-order it for eight thousand
31:44
Josh: dollars which hey for a humanoid robot that can roll around your.
31:47
Ejaaz: House a month or 500 bucks a month so like you know an expensive new york gym
31:51
Ejaaz: membership you can have uh you know this robot walking around and cleaning your
31:54
Ejaaz: apartment with these freaky looking pincers
31:57
Josh: Did you see the demos did it look compelling was this an interesting product.
32:01
Ejaaz: Honestly no it was too slow for me and i hate the the fact that i'm pointing
32:06
Ejaaz: this with my cursor right now that it's effectively one of those old in high
32:10
Ejaaz: school josh did you have those like projectors where like sometimes it
32:13
Josh: Rolls around the.
32:14
Ejaaz: Cart rolls around the cart this looks exactly like that my number one question
32:18
Ejaaz: is like when you come across like a carpet or something like how are you getting
32:22
Ejaaz: over there what like i don't want to have to like be uh
32:25
Ejaaz: the topple monitor where i have to like pick you up if you fall around like
32:28
Ejaaz: how do you know where to place my clothes i I don't know, there's just a number
32:30
Ejaaz: of different questions that doesn't justify $8,000 in my mind, or even $500.
32:36
Ejaaz: But the good news is, this isn't the only company that's releasing robots like this.
32:41
Ejaaz: We had Nori L2 demoed there. Less attractive robot, I have to say,
32:45
Ejaaz: more bare bones, but this is like a demo.
32:48
Ejaaz: It seems to move a little bit more agile and quicker than the previous Weave product or robot.
32:56
Ejaaz: The fact is, a lot of American robotic startups are releasing their products
33:00
Ejaaz: now. And if you compare that to six months ago, it was all theory.
33:04
Ejaaz: It was all hearsay. It was all demos from the likes of FIGURE.
33:07
Ejaaz: You had their robots in the BMW manufacturing plant, if you remember this,
33:10
Ejaaz: and it was supposedly operating all the controls and building cars.
33:14
Ejaaz: Now we see these robots, one, coming at a more affordable rate,
33:17
Ejaaz: and two, being applied to your home.
33:19
Ejaaz: Now the big question is, do you want to allow any of these robots in your home right now?
33:23
Ejaaz: And can this be scaled, mass-produced can this be in my hands right now the
33:27
Ejaaz: answer is still unfortunately no but the target date is a lot closer you can
33:31
Ejaaz: get these robots in a couple of months or by the end of the year at least so you know uh
33:35
Ejaaz: fingers crossed i hope we're entering our robotics era our chat gpt era for
33:38
Ejaaz: robots but uh right now i'm not utterly convinced uh the pinces still throw
33:43
Ejaaz: me off i need some hands i need something like dude i'll
33:47
Josh: Believe it when i see it man ship the product let me see working demos in real
33:51
Josh: people's houses like give me something to really get me invest.
33:54
Ejaaz: In the company you know a few other things like that but only
33:57
Josh: The good ones it's like a lot of most of these companies are going to fail and
33:59
Josh: they're going to just create fun products and they're going to create these
34:02
Josh: little like flashes in a pan and then they're going to dissolve into nothing
34:05
Josh: because turns out it's really difficult to manufacture these things at scale
34:07
Josh: and now they're competing,
34:09
Josh: with the supply chain that the entire planet is also competing on they need
34:13
Josh: the same memory that these other ai companies need and there's only so much
34:16
Josh: of it to go around and the economies of scale are just not really practical for these companies.
34:20
Josh: So in terms of research and development, I think it's amazing.
34:23
Josh: It's fun to kind of hone in on a form factor of what these robots could look like.
34:26
Josh: It's fun to figure out demos and use cases. But in terms of actual practicality,
34:30
Josh: I don't think I'm going to have a humanoid robot in my house this year.
34:32
Josh: I don't think I'm going to have one in my house next year. Maybe 2028,
34:35
Josh: perhaps Optimus will be around by then. Perhaps Figure will have some cool robots by then.
34:39
Josh: But it's going to take a little while to ramp up these production lines and
34:42
Josh: get something that's actually viable via these economies of scale.
34:45
Ejaaz: But here's a question for you, Josh. What would be the one...
34:49
Ejaaz: Pain in the ass that you would definitely allow a robot to solve that would
34:54
Ejaaz: be that would tell you like you know what i'll pay the money i'll pay 300 bucks
34:58
Ejaaz: a month to get this robot in my house to solve this one problem do you have one
35:01
Josh: You know what's funny is like i don't think there is a problem
35:05
Josh: so large that it requires a robot it's like one of the things that's kind of
35:09
Josh: annoying is groceries i have doordash and uber eats they will go and shop for
35:12
Josh: me and they will deliver it or even with whole foods they have their grocery
35:15
Josh: delivery you can get someone to come and clean in your place once a week.
35:18
Josh: And that's like a very good service. And they're going to do a significantly
35:21
Josh: better job than a robot will.
35:23
Josh: Okay i i do laundry but it takes what like 10 minutes to fold the clothes and
35:26
Josh: it's not that big of a deal uh make my bed in the morning when i wake up i like
35:30
Josh: to cook my own food there's a dishwasher to handle the cleaning of it
35:33
Josh: it's like there isn't really anything very pressing in terms of i mean initially
35:37
Josh: my head goes to chores that i think a human or robot would would require so
35:41
Josh: i think that's kind of why i'm waiting just to see,
35:43
Josh: the unique use cases they could figure out or if they just get it down yeah
35:47
Josh: i think it's really down to use cases if there's a killer use case that I see, I'm like, I need that.
35:51
Josh: Then sure, there's no amount that I wouldn't spend on getting this because it's
35:55
Josh: such a cool piece of technology. But there has not been anything even remotely
35:58
Josh: close to that use case. It turns out we've actually automated a lot of things
36:01
Josh: pretty effectively in our life.
36:03
Ejaaz: You see, I've come to the same conclusion, and that's not necessarily bearish
36:06
Ejaaz: robotics. I just think it's bearish robotics in your home.
36:10
Ejaaz: I think robots apply to more industrial use cases, such as manufacturing,
36:14
Ejaaz: or maybe even food delivery makes a lot more sense.
36:17
Ejaaz: I think like fixing that last mile problem, last mile delivery problem where
36:21
Ejaaz: like, you know, you can deliver everything to another country by getting it
36:24
Ejaaz: to that specific address is the most annoying part is a valid use case for robots.
36:28
Ejaaz: I think the manufacturing side of things makes a lot of sense in terms of like
36:31
Ejaaz: car automation or building whatever in the future, maybe even building robots
36:35
Ejaaz: in themselves in the future, which is what Tesla's focused on.
36:38
Ejaaz: This is what Tesla and this is what Prometheus, which is Jeff Bezos's new robotics companies focus on.
36:44
Ejaaz: This is what Atoms, Travis Kalanick's ex-Uber CEO is also focused on as well, right?
36:51
Ejaaz: They're focused on the manufacturing and industrial applications of robots.
36:55
Ejaaz: And I think if you were an investing type of person that wants to focus and
37:00
Ejaaz: figure out where the robotics trends is heading, I think it's in these particular
37:03
Ejaaz: sectors. And that's what I'm
37:07
Ejaaz: I don't know, a robotics specific episode going forwards. If that's something
37:10
Ejaaz: that you guys are listening to and would like to hear more of,
37:13
Ejaaz: I think Josh and I can whip that out. That would be a pretty fun one to do.
37:16
Josh: Yeah, let us know. Would love some feedback. Oh, I wish there was a way to participate
37:19
Josh: in these companies, man. It's driving me crazy. Of all the cool companies we
37:22
Josh: talk about and every single one of them is private.
37:24
Josh: It's like, oh, Etched is awesome. Can't invest. Adams is the most incredible company. Nothing.
37:28
Josh: So I don't know. Hopefully through all of this, we'll figure out a way to get
37:31
Josh: access to the upside of these amazing companies. But I think that's everything.
37:34
Josh: This is a long episode. If you're still rocking with us here,
37:38
Josh: congratulations and thank you. You are a real one.
37:41
Josh: We appreciate everyone making it to the end. The outreach for sponsors has actually
37:45
Josh: been so nice and so sweet.
37:46
Josh: So thank you for everyone who's reached out in terms of getting us help with
37:50
Josh: sponsors, becoming a sponsor yourself.
37:51
Josh: We are looking for people to help support the show so we can keep the lights
37:55
Josh: on and continue to do this for longer. Yeah. So please, if there's anyone you
37:58
Josh: know, please feel free to reach out. We will be happy to converse either through
38:02
Josh: X or in the email in the description below.
38:04
Josh: If you enjoyed this episode don't forget to share it with your friends and if
38:07
Josh: you've made it this far you're all caught up go enjoy the weekend it's fourth
38:11
Josh: of july baby have a great weekend happy birthday america happy semi-quincentennial
38:16
Josh: yeah semi-quincentennial year 250.
38:19
Ejaaz: Years old let's go that's
38:20
Josh: Huge that's huge for the gang like what a country look how far we've come in
38:24
Josh: a quarter of a century or no two,
38:28
Josh: Two and a half centuries?
38:29
Ejaaz: Two and a half centuries. What do you call a thousand?
38:31
Josh: Quarter of a...
38:32
Ejaaz: I actually don't know.
38:33
Josh: An eon?
38:35
Ejaaz: That's a question for Fable. For Fable 5.
38:37
Josh: So when we end this, use your cutting edge AI model to go ask,
38:41
Josh: oh, what is a thousand years? How do we say that?
38:45
Josh: Oh my God, we're so cooked. Happy birthday to us. We are so cooked.
38:48
Josh: At least our brains that we offload to are getting smarter. So even though we're
38:51
Josh: not thinking for ourselves.
38:52
Ejaaz: I officially got smarter this week and I'll be smarter until July 7th until
38:55
Ejaaz: we get to usage-based credits at least at this point.
38:58
Josh: And then I'll probably still wind up using the usage-based credits. But yeah, that's it.
39:02
Josh: Have an amazing 4th of July weekend. Hope to celebrate an exciting milestone
39:06
Josh: for us, EJAS. I'm not sure if you are aware, but we are very close to episode
39:10
Josh: number 200 in Limitless.
39:13
Ejaaz: Which is unbelievable.
39:14
Josh: This right now, if you're listening to this, and also this is news to EJAS too,
39:18
Josh: this is episode 199 of Limitless.
39:22
Josh: So we are one episode away from 200. We're going to record that probably next Monday.
39:26
Josh: And uh that'll be a celebration so for the people who have been here since the
39:28
Josh: beginning like seriously thank you so much it's been an amazing journey thank
39:31
Josh: you um 200 in a little over a year is crazy we're recording yeah four episodes a week and.
39:37
Ejaaz: It's been what 65 000 subscribers on youtube and yeah many many more listeners
39:43
Ejaaz: all over the world across spotify apple music wherever you are waiting commenting
39:47
Ejaaz: letting us know your thoughts uh to go from scratch to this has been insane
39:51
Ejaaz: josh i don't know if you have any any
39:53
Ejaaz: thoughts or highlights about this maybe we should talk about this on the 200th episode we
39:56
Josh: Might have to reflect on episode 200 so stay tuned for that one and i have a
39:59
Josh: feeling it might be about etched because i have a lot of cool things i want to talk about
40:02
Josh: when it comes to that company and inference so as always stay tuned don't forget
40:06
Josh: to share this with a friend if you enjoyed if you find someone else that might
40:09
Josh: enjoy we have a newsletter dropping at the same time you are listening to this
40:12
Josh: as well and that's everything so happy birthday america
40:16
Josh: thank you all for watching and we'll see you next week.
40:19
Ejaaz: See you guys