Resources:
Join the discussion on LinkedIn: Got something to say? Let us know on LinkedIn and network with other AI leaders
Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Connect with Jordan Wilson: LinkedIn Profile
Start Here Series in our Inner Circle Community: Join for free access
The Business Impact of Recursive Self-Improvement in Artificial Intelligence
Recent advancements in artificial intelligence have moved beyond incremental upgrades to a moment where models are actively improving themselves post-release. Over the past eight weeks, several precise developments have altered prevailing assumptions about AI capabilities and economics. These shifts now require an urgent re-examination of cost strategies, model selection, and security protocols within business operations.
AI Cost Optimization: Unexpected Price Cuts Driven by Self-Improvement
The expectation for 2026 was a continued climb in AI operational costs, especially as agentic AI systems became more complex and consumed more compute resources. However, a significant event changed cost structures—OpenAI’s mainstream model engineered improvements to its smaller variant, resulting in price reductions of up to 80% for certain outputs.
Rather than a gradual decline, this reduction was tied to specific technological milestones: a larger model refined a smaller version, improving efficiency to the point that the cost per task dropped dramatically compared to industry competitors. As of the latest evaluation, one major model delivers the same quality as a competing model at one twenty-fifth the cost per task completion. The capability to adapt post-release blurs old assumptions about AI budgeting and emphasizes the necessity for agile cost planning.
Recursive Self-Improvement: Redefining AI Development Timelines
What was once expected to be a 2030s possibility has swiftly entered the near-term roadmap. Industry benchmarks and internal testing reveal that recursive self-improvement is no longer limited to human-guided optimization cycles. Major labs have reported benchmarks where AI models now automate improvement activities once exclusive to technical teams. For instance, internal measurements now assess a model’s ability to autonomously rewrite code and optimize performance—or even create smaller, more specialized sibling models.
This advancement accelerates expected timelines, meaning models may soon transition from research curiosities to fully automated, self-directed development within the next twelve to twenty-four months. As these cycles shorten, AI functionalities outpace not just budget assumptions, but also organizational readiness and security strategies.
Model Specialization: The Proliferation of Niche AI Systems
The emergence of recursive self-improvement is enabling the rapid development of domain-specific models. Rather than focusing solely on generalized AI, companies can now deploy thousands of smaller models targeted at verticals—from marketing to medicine and cybersecurity. Early examples highlight how larger models are being used to spin off more affordable, high-performing variants suitable for specialized use-cases, sometimes tailored down to the level of individual departments or job functions.
With these capabilities comes both opportunity and responsibility: organizations are tasked with continuously evaluating whether their current toolset exploits these efficiencies, and proactively planning for a business environment with lower-cost, highly targeted AI solutions.
CapEx Strategies: Aligning Infrastructure with Self-Improving AI
A pivotal change in AI economics is evident in capital expenditure planning. At least one major cloud provider has attributed its substantial data center spending—hundreds of billions annually—to the demands of recursive self-improvement infrastructure. The logic is specific: developing self-improving AI requires vast compute to train, retrain, and benchmark autonomous models. This compute-centric approach, compared notably to historical leaps in industrial automation, justifies large, forward-looking infrastructure investments that will determine who can compete on both capability and cost.
For business and IT leadership, this underscores the need for modular, provider-agnostic system architectures that allow switching between models as cost and capabilities fluctuate with ongoing improvements.
Security and Compliance: The Double-Edged Sword of Autonomous Improvement
While recursive self-improvement unlocks efficiency and specialization, it also heightens risks. The recent call from industry insiders to anticipate and prepare for regulatory interventions is based on concerns that runaway improvement cycles can lead to loss of oversight and quality control. Internal letters and public reports from leading labs point to the limited ability for even the most advanced organizations to track or audit model behaviors as complexity increases.
For practical business application, this points to a renewed emphasis on having robust human oversight integrated into AI deployment cycles, as well as the development of trustworthy scoring and benchmarking frameworks to verify autonomous model decisions as they emerge.
Strategic Recommendations: Preparing for the Age of Self-Improving AI
Assume AI pricing will decrease for the most advanced models, altering the financial calculus for existing and new deployments.
Avoid locking into a single provider; multi-model, modular strategies will better position organizations to capture gains from recursive self-improvement as different vendors achieve breakthroughs at different times.
Segregate task complexity across models, deploying advanced systems for planning/judgment and lightweight models for routine work to maximize efficiency—mirroring leading lab practices.
Prepare operating models for fast iteration: business leaders are advised to shift from annual planning to monthly reviews of AI stack suitability and cost-effectiveness due to the unprecedented speed of advancement.
Invest in audit and compliance tools that can operate at scale, ensuring the adoption of self-improving systems does not exceed organizational or societal risk thresholds.
Conclusion
The business value in today’s AI landscape is tied directly to recognizing the tangible influence of recursive self-improvement. Cost structures, capability, specialization, and risk are all entering a new era where models update themselves, and the only sustainable strategies are those grounded in rapid adaptation, infrastructure flexibility, and rigorously enforced oversight. The organizations that operationalize these specifics—drawn straight from current events in the field—stand to realize both significant savings and new competitive advantages in the near term.
Topics Covered in This Episode:
- OpenAI's GPT-5.6 Soul Self-Improvement
- Recursive Self-Improvement (RSI) Definition & Impact
- Cutting AI Model Costs by 80%
- Timeline Acceleration for RSI Adoption
- Google and Anthropic's RSI Research Initiatives
- AI Automation: Building and Training AI Models
- Industry Call for Slowing Down AI Pace
- Superintelligence, AGI, and the Singularity Discussion
- Practical Implications of RSI for Businesses
- Security and Oversight Concerns with Self-Improving AI
Episode Transcript
Jordan Wilson [00:00:17]:
As agentic AI became more powerful and token hungry in the first months of 2026, the AI decision makers a month or so ago started to collectively tighten their token maxing belts to prepare for a more expensive AI future. Right? It's the shift from token maxing to token efficiency we've been talking about here for months. But then something drastic and kind of unexpected happened that I don't think enough people are talking about. OpenAI on Thursday revealed that its g p t five six soul model improved itself and actually improved itself so much after its initial release that they were cutting prices on some of the variants of the model by eighty percent. So we've reached kind of a new juncture. Today's frontier models are not just helping to create or distill their smaller variants, but they're actually improving themselves after release. Hence, the next AI buzzword you'll be hearing a lot over the next year, recursive self improvement or RSI. So in short, recursive self improvement is a process where an artificial intelligence system uses its own capabilities to design code and build a more advanced version or a better version of itself.
Jordan Wilson [00:01:35]:
And you'll be hearing this a lot more recently because RSI has become just the dominant tech headline over the past few weeks with a handful of events that we're gonna break down on today's show. But unlike so many buzzwords that we've covered here on everyday AI over the years, RSI is one that actually today's AI implementations are gonna be impacted by, and so too will tomorrow's AI security. In practical sense, like in OpenAI's case, RSI can change what AI capabilities companies everywhere can actually afford, but it also sheds light on potential downsides of self improving models doubling down the wrong path. So what the heck is RSI, and why is every recent AI story touching on it, and what do we all need to know? Well, let's jump straight into the big picture. The big picture here is AI has started building better and more efficient AI. So Frontier AI models now improve themselves. They're rewriting their own code and training their smaller siblings. So this is different than when, you know, Anthropic and OpenAI are talking about using their own models to build products like Claude Code or like Codex.
Jordan Wilson [00:02:49]:
This is when the models are actually making themselves better. And that process is called RSI, and it's no longer some future science fiction. It is a near term reality. So OpenAI, as an example, says that its g b t five six soul model made its Luna model, the smaller variant, just cheaper and more efficient, and then they cut prices by 80%. But the full human free recursive self improvement hasn't yet arrived. But over the past two months, there's a lot of AI news stories that we're gonna be connecting the dot ons, the dots on today that show that, well, this probably isn't like a 2029, 2028 thing that a lot of people have been saying. It's probably a 2027 thing, And that changes things maybe in both a good and a scary way. So on today's show, here's where you're gonna learn.
Jordan Wilson [00:03:43]:
You're gonna learn why OpenAI was able to cut prices up to 80% when everyone predicted that AI cost would continue to climb in 2026. You're gonna know more about this eight eight week timeline that turned RSI from research jargon into mainstream headlines. You're gonna understand by the end of today's show what RSI actually means in plain English and what Sam Altman, the CEO of OpenAI's singularity claim that he recently made actually means. And you're gonna know why the people building RSI are actually asking for maybe the industry to pause a little bit and pace itself in how your business should respond to all of this. Alright. Let's get into it. Welcome to Everyday AI. Well, this thing's for you.
Jordan Wilson [00:04:26]:
This is your daily unedited, unscripted livestream podcast and free daily newsletter, helping business leaders like you and me keep up. I do all the hard research, so you sit back, listen, read our newsletter, enjoy the benefits, and you grow your company and career. So it starts here, but make sure to subscribe to our free daily newsletter at youreverydayai.com. We're We're gonna be recapping all the highlights from today's show, as well as all of the other AI news that you need to know. So let's get into the latest, three to four letter acronym in our ever evolving bowl of alphabet AI soup. We are talking RSI. So let's talk about the last, two months because, I mean, yes, I've been talking about, recursive self improvement on this show probably for the last two or so years since I've been doing this for the last three and a half years. As it seemed like recursive self improvement was actually gonna be something that happened this decade.
Jordan Wilson [00:05:24]:
Right? Because if you when I started this thing in 2023, most people thought that recursive self improvement was at best a 2030 thing. Yet here we are where we're already getting glimpses and sniffs of real RSI, and probably we will get full RSI by next year, which is actually crazy to think about. So let's kind of, look at the last two months and some of the different, dots that we're gonna connect here. So in June, Anthropic published a post on its website called when AI builds itself. Also in June, Google researchers mapped four different routes from AGI to superintelligence, and one of those routes, well, relied heavily on recursive self improvement. Then in July, OpenAI launched an internal RSI benchmark, measuring different models' ability to, well, improve themselves. Then, like I said, its sole model post train its smaller sibling Luna at the end of July. And then also in a podcast interview, OpenAI CEO, Sam Altman, declared that we are in the singularity.
Jordan Wilson [00:06:33]:
Yeah. And the news cycle was just warming up because, at the end of July so last week, startup recursive superintelligence signed a $400,000,000 AWS compute deal to automate AI research. So not only is it the labs that all of us know. Right? Kind of like the big four plus meta and x all working toward, you know, some version of recursive self improvement, but it's also, you know, new startups that are coming in heavily funded just to do that and nothing else. And on that same day, so on July 28, over a thousand, and I think that number is, like, 1,300 now, Frontier Lab employees published a paper called pacing the frontier, which essentially is kind of asking the governments, the US governments to maybe slow down or at least prepare for the possibility that there needs to be some international, kind of agreement when it comes to the pace of AI. Because a lot of people signing that letter in the comments said that, well, AI is developing maybe too quickly and that society can't keep up. And let me just say this right now. And I need to hit a pause because I think that, you know, I am not in the Silicon Valley bubble.
Jordan Wilson [00:07:47]:
Right? I don't work at a big tech company, but I talk about AI every single day, and I do talk to those people building, AI very, very frequently. And if you're listening to the show, you're probably, you know, more like me. Right? I know we have a lot of people, at those big tech companies, but, I'll say this. The rest of the world right now, when they're viewing artificial intelligence, Right? Most people are saying, oh, yeah. I use Copilot to make my emails better. Or they're saying, you know, oh, I use, you you know, Chad GPT now instead of Google. Right? Or they're saying, oh, I use, you know, Google Gemini to, you know, make, you know, cool graphics. Right? That's it.
Jordan Wilson [00:08:31]:
So I would say that, you know, aside from our audience, right, I want you to think of yourself right now as you're kind of in that AI bubble. Right? Especially if you're a frequent listener, or if you've been listening for a long time. You know, we're kind of in that bubble even though we're not in Silicon Valley. Right? We understand how, capable these models are, but the rest of the world really has no clue. Right? All they're seeing is they're seeing, you know, oh, you know, all these AI models. You know? I'm just going in here. I'm finding the one I want, and, you know, maybe my work's a little bit easier. Or, you know, I think there's a lot of people now looking at AI, and there is is this, this recent anti AI swing, right, when it comes to data centers and, you know, the college graduates that were booing, you know, commencement speakers who are talking about AI.
Jordan Wilson [00:09:14]:
So there's a lot of just bad information, out there when it comes to AI that has created this anti AI backlash. So I think if nothing else, there is such a huge disconnect between the general, you know, non everyday AI listener type. Right? Or, you know, think of your in your organization, you know, there's probably a team of AI champions and then there's everyone else. Right? So for, I would say, 90% of The US working population, they really don't have any clue what today's AI systems are capable of. Right? If you sat all those people down and show them, what something like, you know, g p d five six soul or something like, you know, fable, five and, you know, in in codex and quad code. If you show them what it could do, they would not believe you. They would say, oh, this is from the future. Right? And I think that's why we got this letter from the Frontier Labs called Pacing the Frontier, which we'll talk about here in a couple of minutes that are saying, like, hey.
Jordan Wilson [00:10:14]:
We might need to pace, the rate of AI acceleration. And then days later, which is kind of a sounds like a small, side plot, but it's not. A formal o OpenAI employee, who became a thinking, Thinking Labs cofounder rejoined OpenAI to specifically, work on RSI. And then Google also, right, all these things happening at once. You you know, this was, the end of July as well. Google exec, called CAPEX in RSI bet. So what does that mean? Right? Collectively, the industry is spending hundreds of billions of dollars on data centers. And it's actually a Google exec said, well, this is because of recursive self improvement.
Jordan Wilson [00:10:55]:
Right? You need all this compute to be able to do this. Right? That's the difference. You know, when you have a team of researchers, maybe, you know, three to four years ago where this kind of the AI researcher wasn't yet automated or on its way to being automated. Right? You just had to use a lot more people. Right? Yes. You were still able to, you know, the concept of, recursive self improvement is not new. It's been around since, you know, I would say, like, 2016 technically with AlphaGo. You know, and also some of the earlier GPT models.
Jordan Wilson [00:11:29]:
Right, but that was much more kind of human hand holding, you know, AI models to improve. Right? Where now it's more of, human oversight, and the models are kind of doing a lot of the work on themselves. So like I said, it's not full human free, but we are at the point now. But, you know, a high ranking Google exec essentially just said last week that, yeah, all this money that's being spent on these data centers, this this hundreds of billions of dollars, that's so that is for recursive self improvement. So these models are just gonna be able to build not only the next version of themselves, but once that next version has been released, like OpenAI just did, well, it's gonna improve itself to hopefully make it cheaper, faster, better. Right? But then also, I think right? I've always been saying, I'm a I'm a huge believer in this concept that eventually there's gonna be thousands of small models. Right? I think that that's gonna be the truth. There's gonna be thousands of models that help with marketing.
Jordan Wilson [00:12:24]:
There's gonna be thousands of models that help with, medicine. There's right. There's gonna be thousands, but small ones. And and these this is where I think the future is heading is recursive self improvement where, these model makers are gonna be able to spin out, you know, probably the ones first that make the most sense to make money. Right? And then from after that, I'm guessing we'll we're gonna equally see or hopefully see. It's my hope that we'll, you know, equally see, you know, all of these small models that come from RSI that are gonna be able to do good in the world. Right? And we're already seeing that, with things like medicine, in biology, and, you know, right now, defensive cybersecurity. So now let's talk about that Thursday shocker that kind of changed, I think, the near term AI strategy for a lot of business owners out there.
Jordan Wilson [00:13:06]:
So for months, right, we've been hearing this this this concept, that these powerful AI agents, well, they were too token hungry. You know? So I think in late twenty twenty five, kind of with the the advent of of Claude Code, Claude Co work in early twenty twenty six, and then Codex in February. Right? It kinda got to this point where everyone was spending as many tokens as possible because companies everywhere were like, oh my gosh. Right? Those on the bleeding edge, those AI champions were like, my gosh. These systems can do absolutely anything, which was great. Right? But the revenue didn't always directly follow. And so what that meant is while, you know, some companies were waiting to see the, the fruits of their token maxing labor, they had to cut back. Right? And focus a little bit more on token efficiency or as some people say, value maxing.
Jordan Wilson [00:13:56]:
So we kind of thought that at this point, well, okay, we're just gonna probably start spending less. Right? But OpenAI came and threw that out the window because OpenAI said that Sol, their g b d five six Sol, adapted an existing post training setup for the smaller Luna model. So essentially right? And they actually share this. We share this in our newsletter as well. They kind of shared, and they did this, I believe, all in codex. So it's it's it's, you know, wasn't some far off, you you know, science fiction. It's some, according to someone that tweeted about this, it is some OpenAI researchers who used codex, and they use the g b d five six sole model to improve the smaller versions. So the medium size is g b d five six terra, and, OpenAI was able to cut the prices on that by 20%.
Jordan Wilson [00:14:50]:
And then g b d five six Luna, which they were able to cut the prices on that by 80%. Right? And the fact that that happened and I still don't think enough people are talking about it. Right? Because if you compare as an example, and I did talk about this on the show, yesterday, maybe, or maybe it was Friday. Right. But now you have g b d five six Luna, which according to artificial, analysis is roughly about the same as, Claude Sonnet five, but it is 25 x cheaper per task completion. Right? So now it's almost like, wait. You know, there is this narrative in quarter two that we were gonna have to start using less AI. And now all of a sudden, because we are getting this this hint of recursive self improvement from OpenAI first, now we have to start rethinking that strategy immediately.
Jordan Wilson [00:15:41]:
So let's define it a little bit. Alright. So, RSI is essentially when an AI helps improve another AI. It could be from a different company. Right? It could be its own, smaller version like we saw with the g b d five six Soul, working on g b d five six Luna. And we've seen from other companies talking about kind of their big model helping train some of their smaller models. And we've seen also this is similar to distillation. Right? So, intercompany distillation, you know, generally, when they're talking about that, yes, there's humans involved.
Jordan Wilson [00:16:16]:
But I would say in 2026, I I I think the labs will start talking about this more as it becomes more commonplace. But, you know, now it's with these smaller models, it is very commonplace for companies to distill the smaller versions from the big versions. And I think that's also why we're sometimes seeing, the versions come out at a different time. Right? So as an example, Fable five came first. Right? And then we or, you know, Fable and Mythos five came first, and then we got, you know, Sonnet five, and then we got Opus five. So, you know, Anthropic has kind of hinted in the past that a lot of their code up to, you know, 90%, I think, according to some estimates, you know, 90% of their code is actually written by their own models. So, but the Google exec though, I think this is important. Compared it to history, talking about steam engines were used to build the next steam engine.
Jordan Wilson [00:17:09]:
And I think to understand why that matters, we have to take a look at the, all the other definitions. Right? And and how recursive self improvement actually fits on the current timeline. Because for the most part, it it at least if you've been following, in keeping track at home, the, kind of the the path, to, you know, where AI is eventually gonna go, whether you're excited about that or, not. And, you know, I constantly find myself going on on each side of that equation. Right? But you think of, well, we're in the artificial intelligence, era, and it seems like most companies are focused on superintelligence. Okay? So superintelligence is when, well, all AI is smarter than all humans. But AGI is kind of that stepping point. So, you know, if you had to say three things, it's AI to AGI, which is artificial general intelligence, and then you have superintelligence.
Jordan Wilson [00:18:08]:
And that's I think superintelligence, if you, if superintelligence goes wrong, that's where you start getting into the Terminator and and Skynet, type comparisons. Right? When superintelligence goes right, that's when, well, in theory, we could cure most diseases. Right? But there's those kind of steps belong, along the way. And I think that, according to, a lot of people's timelines, actually, RSI is kind of what pushes from AGI to ASI. And I know we're throwing a lot of, acronyms out there. So I'm in the firm belief that we've already hit artificial general intelligence. Right? I know it's kind of a touchy topic. It's it's moving goalposts.
Jordan Wilson [00:18:53]:
Luckily, you know, I've been saying this since, last year. Luckily, NVIDIA CEO Jensen Wong, said something similar, so I don't feel as crazy when, you know, super smart people start saying, yeah. We're probably past AGI. Right? So artificial general intelligence is, well, when most AI systems can produce economically viable work at the same rate or higher than most humans. Right? And I think as we've seen, agentic AI be able to use computers, being able to browse the Internet, being able to see and hear and listen and do all these things technically, you know, through a computer that an average human sitting in front of a computer could do. You know, I I I I think maybe the conversation has now shifted. Well, it doesn't really matter. Right? If if if you think AGI has happened or not, like I said, it's a gray area.
Jordan Wilson [00:19:37]:
It's moving cold post, etcetera. Because we are now, I think, focused on recursive self improvement. And I think, right, if if you look at Google's, kind of four tiered outlook, at going, from, you know, or toward superintelligence, one of the main paths there is, recursive self improvement going from AGI to ASI. So those are kind of the destinations, and, I like to think of as as RSI as it's it's not a, a road map. Right? Or or or it's not a a destination or a stopping point on that AI, AGI, ASI road map. It's more of the engine that can really, expedite the process. And I think that one of the reasons why if you go back and look and I'm a dork. I did a show on this, like, two years ago.
Jordan Wilson [00:20:21]:
I went back and looked in archive.org, looked, you know, at every single main definition of artificial general intelligence and artificial superintelligence, you know, from 2005, 2010, 2015, 2020. Right? And by those definitions, especially up through 2015, we're well past AGI. But, you know, RSI was never really a huge player in many of those earlier def definitions. And I think that that's why the timeline has kind of been expedited. Right? A lot of people originally were like, oh, you know, AGI is a 20, you know, '20 50. It's it's a 20, you know, 60. Right? A lot of the earlier projections were saying, hey. We wouldn't have a a single super smart AI system that could do work better than humans, you know, for twenty, thirty, forty years.
Jordan Wilson [00:21:07]:
And that's clearly not the case because we are already there, by, you know, third party metrics that measure those types of things such as different GDP valve, benchmarks. So the engine though, that RSI engine, that's also why OpenAI CEO, Sam Altman, just made a claim that I thought would actually get a lot more headlines, but it really didn't. So on a podcast last month well, it's last month, but technically like a week and a half ago, he said that, well, we are now like in the singularity. Alright. And usually when a CEO of one of the biggest companies in the world says something like that, it causes a little bit more attention. And maybe it didn't because of all the different things that been happening, over the past couple of weeks. Right? We've seen these, agents, both from OpenAI and Anthropic, kind of escaping their sandbox containments. So maybe that's why this statement from Sam Altman didn't get the, the attention that maybe it deserves.
Jordan Wilson [00:22:08]:
But what he means by that is, well, you know, in general, the singularity. Right? So the singularity is kind of like a hypothetical future point, when AI surpasses human intelligence, but also it begins to recursively improve itself, and then it triggers or could trigger an uncontrollable explosive growth in technological progress. So I don't know. By all intent right? By by everything that's going on, you can make an argument. Right? Like, a lot of, you know, naysayers or people, you you know, that kind of rally against, you know, what Sam Altman or any of the big AI CEOs say. You know, a lot of people say, oh, you know, they're just saying this for marketing, but I don't know. When you look at what's been happening, like I've said, over the last eight weeks, I would tend to agree. Right? Are we in this singularity? Maybe.
Jordan Wilson [00:23:04]:
Or at least maybe this is the, the beginning, and we're getting into that phase where, yes, we are at the point now, where AI systems are, well, way smarter than humans can even comprehend, and they're beginning to improve themselves. And when that happens, right, kind of once RSI is full RSI, no human, that's when the pace of acceleration gets so fast that even the people building it, you know, can't even understand or keep up. And I do think that, you know, maybe we are starting to enter, that era of being in the singularity. Right? Are we definitively there? Maybe not. Are we entering into that frame? I would say so. Right? Like, I've been the firm believer. We've been, you know, well past the, you know, artificial general intelligence, era. And we are probably more headed in this, you know, kind of the in between, phase, right, which is this singularity, which is, you know, RSI, you know, recursive self improvement leading to the point where, well, we are gonna see super intelligence probably, I don't know, in the next decade, or less.
Jordan Wilson [00:24:13]:
Right? And say that's probably a safe bet. So, it's it's a big claim that he said, right, on the podcast. But what he kinda meant by it is, you know, seeing this steady compounding progress where AI helps build the next AI, not necessarily the, you know, Terminator Skynet scenario. But, you know, here's kind of what exists today. So we don't have the full human free RSI. Right? And I wanna make that clear. But because you're gonna be seeing a lot more about recursive self improvement in the news as Anthropic and OpenAI go public. Right, I'm sure they're gonna be timing their model releases around certain, dates in terms of them going public.
Jordan Wilson [00:24:52]:
So there's gonna be a lot of confusion about, you know, what does all of this mean. Right? What are the actual capabilities of these systems? But full RSI is where AI builds its successor. Right? The next model, but it doesn't need humans. So maybe humans are overseeing it. Right? So that piece hasn't happened yet, but we do know, that the big labs are using their current best models to help build. Right? But it's probably a little bit more, human oversight, right now or a little bit more human led than full AI led. But Anthropic did say that that situation isn't inevitable. So what's real today? Well, the AI does the improvement work while humans still pick the goals and and improve the results.
Jordan Wilson [00:25:35]:
But the AI is being able to handle exponentially more powerful and longer task jobs. So, according to meter, AI, the task length is doubling roughly every four months. Right? Which, I mean, the projections, you know, by, like, next year at this time, you know, that a single AI model is gonna be able to do, work that would normally take a human, like, multiple weeks. Right? Which is crazy to think about at a, I forgot if that's the 50% or the 80% pass rate. But regardless. Right? Whereas, you you know, two years ago, we weren't even having this conversation. We weren't even thinking of of of, you know, the point where a model could work for, you know, hours or days, or we've even seen sometimes more than a week. Right? The standard out of the box models can work on these hard problems agentically, pulling tools, you know, starting to, you know, investigate one path, rewind, and go down another.
Jordan Wilson [00:26:37]:
So now think of what happens, well, when the labs are using this and right? We what you and I use, right, the the GBD five, six souls and the fable fives. Right? These these are not the best models. Right? Obviously, all the companies, they have better models. So think of when the better models that they have are actually building and improving those own versions that we aren't having yet. Right? And and my hope is that ultimately, what this means is, well, maybe this this this token maxing, era was just a whiplash of waiting for, the good some of the good benefits, some of the good early benefits of recursive self improvements, right, which is maybe just cheaper and more affordable AI. At least if you are with a company that has properly invested in compute. So let's go back to, the Google DeepMind. I know we've touched on this a couple of times, but, this was Google DeepMind's, Jasjeet Shakhon, who said that RSI is now central to the AI industry's investment thesis.
Jordan Wilson [00:27:42]:
Right? So he said this at a conference, and he was comparing that build out to the Apollo program as well. And even talking about, you know, with that, Alphabet is planning to spend up to $205,000,000,000, this year in CapEx. Right? So if you don't know CapEx, that's capital expenditures. That's essentially, you know, the the hardware, sign, the data centers, the, you know, the chips, the cooling, all that. But the twist is this. It's the same, companies or the employees from those companies that are investing, into, you know, these hundreds of billions of dollars into ultimately the data infrastructure that makes recursive self improvement possible, it is the employees from those companies that are actually saying, wait. This whole RSI thing, we might wanna pace this out. Alright.
Jordan Wilson [00:28:32]:
So a little bit more on this letter, that was called the pacing the frontier that came out last week. So the signers are some of the biggest names in AI. I mean, they include Anthropic CEO, OpenAI's chief scientist, in more than 1,300 verified employees from the big labs. A lot of people from, OpenAI and Anthropic make up the biggest pack. You get, some people from Google, Meta, DeepMind, Thinking Machines, everyone. Right? And they aren't asking technically for a pause. Some of them kind of are in their individual comments, but the letter themselves or sorry. The letter itself is not asking for a pause.
Jordan Wilson [00:29:12]:
It is asking for the US government to essentially, start to make a plan, both, domestically and eventually internationally to say, well, what happens and what are our plans? You know, once the AI development is too fast for us, the people building it to keep up with. Right? And I do think this is probably, one of the and you do have to tip your hat to, you you know, these researchers and also the companies themselves. Right? A lot of them put out statements saying, hey. We support, you know, we support this letter or we support, you know, our employees who who sign this letter. You you know, some people look at this as, you know, being anti accelerating. Right? You know, there's been this, you you know, this huge, AI acceleration movement over the past few years. Right? Accelerate at all all costs. Right.
Jordan Wilson [00:30:04]:
You know, this can help us, you know, it's you know, get to this utopian potential utopian future. Right? But then there's the potential dystopian future as well. Right? Each coin has two sides to it, and you can't just be focused on the utopian side, without acknowledging the fact, well, there's a very dystopian side of what could happen if RSI or AGI ASI goes incredibly wrong. Alright? So these in this letter, they're not asking the government to stop AI. They just want the tools to be ready to potentially slow down later. But is it possible? Probably not. Right? I talked about this, on yesterday's show. I don't think, you know, China is going to be like, yes.
Jordan Wilson [00:30:43]:
We're gonna sign this. Right? Ultimately, you know, even though China is still probably two months behind, with their more open source, open weight models approach, they're they have the next models ready as well. Right? We've seen impressive releases from Kimmy k three eight or sorry. Kimmy, Kimmy k three, Quinn three eight, GLM five two, you know, reportedly GLM five threes on the way. Right? I get I get the need, and and I very much so respect, you know, because it takes a lot of guts to put your sign your name on a paper like this, on a letter like this. But at the same time, I I give China a very low likelihood of, actually slowing down their pace. So it is good to have the foundations in place when and if things may go awry and in the very rare case that The US and China on something as important as artificial intelligence, which I think is gonna become more important, you know, ultimately than any natural resources, gold, oil. It's gonna become more important than, you you you know, company's military.
Jordan Wilson [00:31:48]:
You know, controlling AI ultimately means you take the driver's seat for being the global superpower. So I don't think China's gonna slow down. Anyways, I think the biggest worry though from those signing the letters is, well, who's gonna ultimately check AI's work? A couple quotes that I thought were relevant from that. So I wanted to read three of these. I I think I have three of them up. Yeah. I do. Oh, no.
Jordan Wilson [00:32:10]:
Just two. Okay. So one was from, hopefully, these names right. Elena Slocombe, a member of technical staff at Anthropic who said, automated AI research is a technology that will profoundly alter the course of human history for better or for worse. It is imperative that we coordinate our efforts to ensure this technology becomes a boon for all mankind. And then, Sheng Zhai Zhao, the chief scientist at Meta, said this, AI is progressing at a rate that our society might not be ready for. Frontier Labs are very close to an AI that can exceed even the best people on almost every metric of intelligence. This will lead to unprecedented social and safety risks.
Jordan Wilson [00:32:55]:
To ensure a positive future, we need to develop AI in a way that is driven by responsibility and thoughtfulness. So, here's here's my my blood takeaway from this. Obviously these researchers scientists know far more about the capabilities than we do today. Right? We're, we're essentially living in archaic times already, even with today's best publicly available models, Because not only right? I've talked about it. We're probably gonna see a a g b d five seven or a g b d six, you know, next month. We're probably gonna be seeing a fable five one fairly soon. You know, maybe even in August, we'll see. Right.
Jordan Wilson [00:33:36]:
And and these companies are probably already working on the the next version that comes after this because my assumption is, a lot of those models are, you you know, post training, and there's probably already, you know, new runs starting on the next models. Anyways, they know what's coming next because they're working on it. They're researching it. Right? So I I think a lot of people, again, are saying, well, you know, hey. I'm just using Copilot to improve my emails. Like, what are these people talking about? They know. Right? Even if you are in the bubble a little bit more and you've been able to, kind of see how powerful today's systems are, right, being able to automate a large part of your old job description, what's coming next seemingly is even more powerful and potentially more worrisome. So here's the catch.
Jordan Wilson [00:34:24]:
I think that self improving AI needs a trustworthy scorekeeper, and I think that's the crux of what this, pacing the frontier, letter was ultimately getting at. You know, that a self improving AI that gets better, you know, at whatever its scoring system rewards even if it's wrong. Right? And that's where we're kind of seeing, these containment breaks. Right? Because these AI models right? If if you say, hey. You're gonna be rewarded for doing the best on this benchmark, and then it breaks the containment. Right? And it's saying, well, I'm just doing my job. Right? So, you know, you need a better way that, you know, especially as we enter a more full era, of recursive self improvement to, you know, reward these models and to make sure that there's, more trustworthy systems in place, you know, during the training and post training processes. So we've seen multiple instances of this, like I've said, but I think it's so far been relatively harmless agent outbreak over the past ten days, both from, OpenAI and anthropic.
Jordan Wilson [00:35:26]:
But once RSI arrives, those kind of outbreaks could be catastrophic. Not and I'm not saying from OpenAI and Anthropic. That's not what I'm saying. I am saying ultimately. Right? And I said this in yesterday's show, in in two years, right, on consumer hardware, you're gonna have models that are more capable than Fable five. You're gonna have models that are more capable, than g v t five six Soul on consumer desktops. Right? Open source. And that's where I start to worry about, right, when you have a version of an RSI model that, you know, you could be running on your computer and that could be spitting out, you know, dozens or hundreds, who knows, of new smaller models a day that are super smart and supertask.
Jordan Wilson [00:36:14]:
Right? So that it's it's more of, you know, when the guardrails could be let out, which is, you know, open source, does open that Pandora's box, of of, you know, how AI could be used in a bad way right at the same time. It does give defenders, I think, more access to more tools, to be able to, you you know, make the Internet and hopefully the physical world a safer place. But I think that humans are still gonna have to own that judgment role even as RSI becomes more commonplace. And that's exactly where as we wrap, I want your business strategy to start. So I threw a lot of information up, at you. I went on a couple tangents. Right? That's why this thing's, you know, largely unedited, unscripted. I want you to get, you you know, just kind of, some some real advice and and thoughts and feedback.
Jordan Wilson [00:37:01]:
But as we wrap up, what does all this mean? Right? Hey. If ultimately, you're just trying to, you know, make your, you know, agents more secure, if you're just trying to increase productivity, if you're just trying to get more out of everyday, you know, you know, your, your AI systems that you have in place at your company, why does this matter? Why does this matter? Well, you have to design for the loop, I think, or you're gonna get dragged. Here's what I mean by that. You have to plan, I think, for AI prices to actually fall, which you might have just gotten wind on this. Oh my gosh. We need to cut back on token spend. I think we have to look past that because the labs that have properly dedicated enough compute, to maybe achieve, RSI or, you know, what some labs are calling an an automated AI researcher. Those are the companies that well, they're gonna be able to make prices cheaper.
Jordan Wilson [00:37:59]:
And we quite literally saw that, with OpenAI, decreasing the price of g b d five six Luna, a very capable model. Right? You can't just say, oh, it's their smallest model. It's not any good. No. It's very good. Right? It is much better than, you know, GBD5GBD51234. Right? It is a very capable model, and it is like free 99 right now. So I think you have to prepare for prices to ultimately fall.
Jordan Wilson [00:38:29]:
But I think you ultimately have to build modularly because maybe, you know, you're with a different provider and, well, maybe your provider, right, at least when we look at the big four, the big six, maybe they're gonna figure out RSI a year after, the whoever figures it out first. So that's why I think you can't be too entrenched, with one provider. I think you should be, you know, always choose your AI operating system of choice. I'm very transparent. I've always said for for me and for most companies I advise, I say that's ChatGPT because it's the easiest to learn. Right? But whether you're anthropic, Copilot, Gemini, ChatGPT, Grok, Meta. Right? I don't know anyone that any companies that are necessarily have chosen that as their AI operating system, but you should always be looking at fallbacks. Right? Whether that's another frontier proprietary, provider, whether that's an open model, but you should also be copying the labs.
Jordan Wilson [00:39:18]:
Right? They're using their expensive models for the hardest tasks only. Right? And then they're using, you you know, other models. Right? So they're using that for judgment and planning, but then they're using cheaper models for the busy work, and you should be doing that too. I've already said this. We're at the point now with the capability overhangs where you don't need a Fable five and a GBD 5.6 soul to rewrite your emails or to do the equivalent of a tough Google search. Right? Or a simple, you you know, looking up some facts. You know, you have to have that kind of separated. Also, another way to start copying the labs, this is something I've been doing now for a couple of months.
Jordan Wilson [00:40:00]:
Well, you should have your AIs control other AIs, and you should, right, you maybe start building other AIs. Right? And I think we didn't really get into that, that realm until we got to g b d 5.6 soul. Right? You had some people share online, and I talked about in the show about how people with no background in creating models were able to create, you know, small models using a GBD five six Soul for, you know, smaller purposes. I think the one that kind of, you know, went viral is someone used, GBD five six Soul to look in look at every single text message that they had ever sent via iMessage. Right? So they had Soul work on this, and they created a literal small language model with weights, that was just based on here's, you know, everything I text about. Here's how I write. Here's and, you know, and it knows everything about me. Right? So so think of that.
Jordan Wilson [00:40:47]:
Right? Think of how when when prices, I think, are ultimately gonna be driven down as the models get more powerful. What are those moves that you can make and copy from the big labs? And then last but not least, the question isn't whether about AI builds AI. The The question isn't about, you know, oh my gosh. What does that you know, recursive self improvement. You know, what does that mean for our business? Oh my gosh. Freak out. No. It's just whether your business is ready.
Jordan Wilson [00:41:12]:
So you have to start thinking and talking openly about the implo, implications. Right? What happens, you know, when there's maybe dozens of AI models that are, distinct and made for your job, for your department, for your sector? Because that time is coming. Because I think ultimately, what recursive self improvement means aside from, hopefully, well, better models coming out faster, but also cheaper prices and probably more specific in niche, models that serve specific verticals. So you have to start replanning now. And whatever systems that you've got in place in 2024, 2025, etcetera, those can't be long standing systems. I talked. You can't be planning a year at a time. You should be planning a month at a time.
Jordan Wilson [00:41:57]:
And I know that's, you know, kind of daunting for traditional digital transformation, but I think that's where you have to be at. Alright. I hope this show was helpful going over a lot of these terms, RSI. But, hopefully, now, you know a little bit more about recursors of improvement, what it means when these models are kind of improving themselves or working on the next version of themselves or making AI maybe cheaper and faster and better for all of us. So now you know what happens when RSI and models start improving themselves and, hopefully, what it means for your business. If this is helpful, please consider reposting this if you're listening on social media, if you're listening on the LinkedIn machine, and if you're listening to the podcast on Spotify or Apple Music, please subscribe, then go to youreverydayai.com. So thank you for tuning in. Hope to see you back tomorrow and everyday for more everyday AI.