Sep 30, 2026 · 46 min · 10 segments
I realize I am dating myself to some extent, but my first cell phone plan came with a fixed number of minutes and a small allotment of text messages, and anything beyond that was billed one at a time…
Richard LunaGuestTechSpective Podcast HostHost
But the amount of money that's being borrowed to build all the data centers they are not they're 20 a month is a joke it's going to be 2 000 a month and when you get to that level of cost and you can say well they'll get cheaper right over time not so much i mean i'm waiting for my a new macbook to come which has got gobs and gobs of memory so i can run um a local model because my current one i i I burn up all the RAM on it.
But when I run a local model, I'm getting 90% of what I can get on the public network.
So as we get machines that are better and faster, why am I going to use the public network for my AI where everybody can share it and see what I'm asking and they can get smarter? Why do I want my world exposed and have all the data out there? I still want to benefit from the processing abilities of AI, but that's my take on it.
It's a whole series of vectors, right? It's security, it's privacy, it's cost.
As AI deploys more and more, how do we train people? Because I don't need an admin if I can have bots do what I need the bots to do.
Right, right.
But I know what
Right.
I mean, that's a whole ongoing, you know, like you said, there's many layers to it because there's no one scenario and there's no one right answer.
First, I was going to say, it is interesting to me, and I know people talk about this all the time.
They talk about how, like, you know, we managed to, you know, go to the moon and back with, like, you know, less memory than a pocket calculator could.
My current Surface laptop has 64 gigabytes of memory.
And yet, because everything we're running and everything we're doing is also exponentially using more resources than it did the net experience has never really improved.
The net experience kind of stays the same or maybe even gets worse over time, even though the resources have gone up, the resource consumption is going up faster.
Yep.
So I did download a local LLM to run on here.
But the amount of money that's being borrowed to build all the data centers they are not they're 20 a month is a joke it's going to be 2 000 a month and when you get to that level of cost and you can say well they'll get cheaper right over time not so much i mean i'm waiting for my a new macbook to come which has got gobs and gobs of memory so i can run um a local model because my current one i i I burn up all the RAM on it.
But when I run a local model, I'm getting 90% of what I can get on the public network.
So as we get machines that are better and faster, why am I going to use the public network for my AI where everybody can share it and see what I'm asking and they can get smarter? Why do I want my world exposed and have all the data out there? I still want to benefit from the processing abilities of AI, but that's my take on it.
It's a whole series of vectors, right? It's security, it's privacy, it's cost.
As AI deploys more and more, how do we train people? Because I don't need an admin if I can have bots do what I need the bots to do.
Right, right.
But I know what
Right.
I mean, that's a whole ongoing, you know, like you said, there's many layers to it because there's no one scenario and there's no one right answer.
First, I was going to say, it is interesting to me, and I know people talk about this all the time.
They talk about how, like, you know, we managed to, you know, go to the moon and back with, like, you know, less memory than a pocket calculator could.
My current Surface laptop has 64 gigabytes of memory.
And yet, because everything we're running and everything we're doing is also exponentially using more resources than it did the net experience has never really improved.
The net experience kind of stays the same or maybe even gets worse over time, even though the resources have gone up, the resource consumption is going up faster.
Yep.
So I did download a local LLM to run on here.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.