Skip to main content
Matt Hicks

Matt Hicks

CEO of Red Hat

Oct 20, 2026

31:24
Could you make the case that open source genuinely beats the proprietary models or could you address where maybe there's some opportunity for growth for them?
31:34
I think there are two.
31:35
In my case that I shared, we started with all frontier models.
31:39
They're great.
31:40
They're incredibly capable.
31:41
We use all of them.
31:44
But when you get into a domain and a task, the big models are good at everything, but they can be beaten by models that maybe don't have the same world knowledge but have enough to do my job really well.

12 MINS LATER

44:23
Were there levels of confinement that Red Hat, first of all, put in, obviously, but what did you decide to leave out? That's more interesting when you're talking about getting contributions.
17:52
How do you tell, uh, an enterprise as to, you know, what's your framework of building that infrastructure that it becomes model agnostic?
18:03
Yeah.
18:04
Th-this one, my recommendation is the, as much as we love small models, and let's-- I'll put models into three bands of it, overly simplified, but Uh, when we talk parameter counts, you have these 10 billion parameter models.
18:22
Um, they're big for computing, but very small for AI models.
18:26
They can do a task of work that is very hard to program, uh, but it, it creates new opportunities to do simple things with it.
18:35
What they don't do is they don't reason.
18:37
They can't think through complex problems.
22:08
So perhaps at least to start off with, could you give us a back-- uh, you know, an understanding of the concept itself of what it does? And second, is there a parallel to that in the AI world as I'm, you know, embedding more AI across my products as I have to do a similar, uh, framework?
speaker_0HOST
105:09
Mm-hmm.
105:09
... to answer those questions you're gonna ask it.
105:12
And so we work with customers to really find that use case.
105:16
It doesn't always fit every aspect, but how many of those pieces can you move to something that is a hundred times cheaper with it? And it requires some, you know, you're competing with the Googles of the world or Microsofts in terms of operational discipline, so you have to be able to run these things well.
105:36
Um, this is word you hear, inference, that pops up.
105:39
We help enterprises with inference.
105:41
Just normal, mere mortal data centers, GPUs, NVIDIA, AMD, and running these models efficiently so you can start to decompose that problem into, you know, what are the areas we can optimize and really control ourselves.
speaker_0HOST
106:31
Uh, but, but how do those conversations go specifically, and how much spend today on Frontier models really shouldn't even be going to Frontier models and should be going to, uh, some of these smaller, uh, more, more, um, uh, s- uh, specific models?

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.