Skip to main content
Scott Guthrie

Scott Guthrie

DeveloperWikipedia

Search complete. 6 mentions across 6 episodes found for "Scott Guthrie".

Oct 14, 2026

speaker_1HOST
6:25
Chang would have sent this filing back for that same kind of security.
speaker_1HOST
6:29
Scott Guthrie on the official Microsoft blog.
speaker_2SOUNDBITE_SPEAKER
6:33
As we move into the era of superintelligence, Microsoft is entering a new phase of infrastructure.
speaker_2SOUNDBITE_SPEAKER
6:39
We will scale our Azure Maia family to multi-gigawatt deployments starting in twenty twenty-seven, making Maia the primary inference system for Microsoft products and services.
Deirdre BosaGUEST
9:14
And on that note too, I would just say Microsoft did something really interesting yesterday.
Deirdre BosaGUEST
9:18
They announced the launch of Maia 300, their new, you know, custom AI chip, and I was talking to Scott Guthrie, the cloud chief at Microsoft, and he brought up something that I hadn't thought about.
Deirdre BosaGUEST
9:29
Like, yes, when you have your own chips, you can reduce your reliance on Nvidia and AMD.
Deirdre BosaGUEST
9:35
Maybe that helps your margins.
EliHOST
13:55
Leaders cut back on the company's Cloud Code subscriptions.
EliHOST
13:57
Senior leaders, including Scott Guthrie, who runs Microsoft Azure Cloud Unit, and Jay Parikh, who runs Microsoft Developer Tools, told staff to be more cautious of their coding use and urge leaders to monitor how much their employees were spending and to prioritize non-anthropic models.
EliHOST
14:15
So basically, the idea here is they're saying, hey, take a look at how much you're spending.
EliHOST
14:20
Oh, my God.

Unknown podcast

Ep. 20 — Agents, Chips, Balloons: The AI Stories Reshaping 2026

Oct 13 · 1 Mention

AdamHOST
2:30
Built on TSMC's N2P process with eight HBM4 stacks, it offers eighteen terabytes per second memory bandwidth per chip in racks with a one hundred and sixty-seven kilowatt power envelope.
AdamHOST
2:42
Production deployments are set for 2027, with multi-gigawatt scale plans announced by Scott Guthrie.
AdamHOST
2:48
Meanwhile, the Maia 200 is already running production workloads for Copilot and Teams, delivering up to one point seven times more tokens per watt than GPU infrastructure.
AdamHOST
2:58
Evercore ISI estimates Maia could represent only about ten percent of Microsoft's AI chip spending by early 2027, so Nvidia still dominates near term.
Brian McCulloughHOST
5:50
Quoting The Verge, "Maia 300 not only improves the cost and efficiency of running AI workloads in the cloud, it's also more capable of handling the latest reasoning models and agentic tasks." Maia 300 will start being rolled out to production in twenty twenty-seven, and it has been designed with efficiency in mind.
Brian McCulloughHOST
6:09
"To-- we expect Maia 300 to deliver forty percent more tokens per megawatt compared to leading GPU-based systems," says Scott Guthrie, executive vice president of Microsoft's Cloud Group.
Brian McCulloughHOST
6:20
"And in a full ninety-six accelerator configuration, it delivers up to seven x the rack scale token throughput of Maia 200." These efficiency improvements will allow Microsoft to rely mainly on its in-house AI chips soon, instead of ones from Nvidia and AMD.
Brian McCulloughHOST
6:39
"We will scale our Azure Maia family to multi-gigawatt deployments starting in twenty twenty-seven, making Maia the primary AI inference system for Microsoft products and services," says Guthrie.
speaker_0SOUNDBITE_SPEAKER
0:05
Hi, I'm Adam Fowler.
speaker_0SOUNDBITE_SPEAKER
0:06
Hi, I'm Scott Guthrie, and you're listening to the Need to Know podcast.
speaker_0SOUNDBITE_SPEAKER
0:09
All the latest Microsoft Cloud news, as well as industry guest deep dive conversations.
speaker_0SOUNDBITE_SPEAKER
0:15
It's the Need to Know podcast.

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.