Cloud Wars Live with Bob Evans

In today’s Cloud Wars Minute, I look at how Google Cloud is combining AI agents, enterprise data, specialized skills, and partners to transform industry-specific solutions.
Highlights
00:04 — Over the last few years, Google Cloud has made different moves to address what I believe is an enormous opportunity in industry-specific solutions. It's gone about this a few different ways, always leaving the door open to work very closely with partners, whether that's integrators or ISVs. And, I think it's on to now the right way, as they have now gone industry-specific with Gemini Enterprise.
00:32 — It's putting around what Thomas Kurian calls a system that involves skills plus data plus agents plus partners moving forward on this. It's probably going to go across a lot of different vertical industries. The first two they're addressing are financial services and legal. Kurian said general-purpose AI is terrific, but it's just not enough to be able to handle these industry-specific solutions.
01:34 — And, one of the things he said here this is only the beginning. That could mean going deeper into financial services and legal, but also expanding broadly across other industries. It's laid the foundation: skills that can be encoded into the software, access to relevant data, agents that can get real work done and complete projects, and an ecosystem that’s completely open to partners.
03:08 — Kurian said, "We put together nice scaffolding for partners on which they can develop specific solutions." And he said, "We also make it very easy for those solutions to be discoverable by customers or prospects," and how this has worked together. He kept emphasizing, "We want this to be simple, easy."
04:37 — And I think, paired up now with the hugely successful Gemini Enterprise, it's going to give businesses now looking to move into AI — not just in a broad sense, but for industry-specific capabilities — I think this is going to be a big boost, and I think this is going to be a model that other vendors in the Cloud Wars Top 10 follow.

Visit Cloud Wars for more.

In today's Cloud Wars Minute, I look at how Microsoft is bringing Arabic-language AI models into its ecosystem through its strategic collaboration with HUMAIN.
Highlights
00:09 — Microsoft has announced a long-term strategic collaboration with HUMAIN. They're a Saudi Arabian AI company owned by the Public Investment Fund. That's Saudi Arabia's sovereign wealth fund. So, what we're looking at here, and this is why I think it's particularly interesting when we're analyzing Microsoft's strategy, we're looking at something that isn't necessarily an infrastructure deal.
00:35 — We've seen quite a few of those recently. Instead, this is more of a deployment and go-to-market situation. So, the crux of this partnership includes a number of steps from both sides. HUMAIN will bring its Arabic-language ALLaM models to the Microsoft ecosystem through Microsoft Foundry.
00:54 — Those models will also be integrated into Microsoft 365 Copilot, which will allow companies to create specialized agents using ALLaM for Arabic-language tasks. And beyond this, HUMAIN's AI specialists are going to work alongside Microsoft forward-deployed engineers with customers to really accelerate AI from experimentation to production.
01:18 — Now, further down the line, the two companies are going to explore deeper ties and use cases, including productivity, devices, models, perhaps infrastructure, and joint go-to-market initiatives. Now, this is, of course, going to start in Saudi Arabia, but both companies can see these efforts expanding across the broader Middle East and Africa region.
01:43 — I think it really highlights the importance of not just seeing AI as an English-language technology. And by explicitly focusing on bringing Arabic-language AI models into its ecosystem, Microsoft has positioned itself to build a much larger and powerful presence in that region.

Visit Cloud Wars for more.

In today's Cloud Wars Minute, I analyze how Aneel Bhusri is reshaping Workday’s agentic AI strategy following his return as CEO.
Highlights
00:03 — Interesting conversations on the Workday FY27 Q2 earnings call last week. One of the things that came up here was the difference in their approach to agentic AI that's come about over the last several months since the return of CEO Aneel Bhusri.
00:48 — One of the big things that happened here, he said, we don't necessarily want to have a whole lot of agents for a lot of simple functions. We want to focus on agentic AI that targets big, complex processes and challenges that our companies, that our customers are facing in HR and financials.
01:08 — Here's one of the quotes that he gave on the Q2 earnings call. He said, "We have to have agents that deliver real ROI." He said, "We've killed a lot of agents since I've come back, or we've rolled small, simple agents up into bigger, more complex agents."
01:27 — He said, "Anybody can solve the simple problems. That's not a place where we're going to be able to deliver this real ROI for customers." So, interesting time now for Workday, right? Because they're a big company, right? They're guiding to $10 billion in revenue for FY27, which will end January 31st.
02:41 — I've got an article later that offers some more perspectives on this. Also, some comments from President and Chief Product and Technology Officer Gerrit Kazmeier. He goes into some detail about why this new wave of agents is so vital: that they've got to be able to be connected to everything, be able to handle these complex problems, and drive that big ROI.

Visit Cloud Wars for more.

In today's Cloud Wars Minute, I examine why Meta is buying massive amounts of AI capacity from Microsoft despite building its own AI infrastructure.
Highlights
00:10 — Meta has become one of Microsoft's largest AI customers, spending hundreds of millions of dollars annually through Azure and consuming trillions of tokens per week. It's a massive figure.
00:24 — But to me, the real story here is about how AI is utterly transforming relationships in the tech world. Even though Meta is building its own AI infrastructure, it's still buying massive amounts of outside AI capacity through Azure, and the stakes are so high in this race that no company can afford to slow down.
00:47 — Even if that means making a competitor richer in the process. It's also a massive validation for Microsoft's Azure AI Foundry strategy, which means Microsoft can make money regardless of which AI model comes out on top. So, AI companies — and I'll use that term, AI companies, because really that's what all these tech businesses are becoming — AI companies are increasingly becoming customers of one another.
01:13 — And I think we've sailed past the "who builds the best model wins" competition, and now it's about something else. It's about infrastructure. In the battle for supremacy in this AI race or evolution, it may not be who builds the best AI, but who owns the best infrastructure to run and access it that will come out on top.

Visit Cloud Wars for more.

In today's Cloud Wars Minute, I ask whether OpenAI has quietly abandoned its AI super app as customer confusion around AI continues to grow.
Highlights
00:02 — I wonder if over at OpenAI it's got a detective case going on here — the case of the missing AI super app. This is a concept, an idea that OpenAI floated very visibly, very demonstratively five months ago at the end of March, when the company put out a landmark blog post outlining its valuation.
00:44 — It laid out in detail its plans, its strategy, how, you know, compute is the base, then models, then that leads, you know, to other steps, all of which makes businesses and individuals more capable.
01:48 — But, in the August 24 Sarah Friar post on LinkedIn, there was zero mention of this AI super app. Now, was it just an oversight? I don't think so. Sarah Friar's pretty good with the details, pretty good with the strategy, and I wonder if, in hindsight, they're saying, you know, customers are already being overwhelmed with so many new things.
02:47 — What seemed like a great idea five months ago might not be such a great idea now. They've perhaps got other priorities, and I just think this notion of calling something a super app comes at a very bad time. I've got a detailed article later this morning on CloudWars.com where I offer a lot of context, and I compare side by side.
04:44 — I think, therefore, that it's possible that OpenAI has rethought this and said, "We're going to blow by this AI super app thing. It's just too much novelty, too much of a distraction, perhaps too much confusion for customers here to deal with when they have so many other highly relevant, urgent things to grapple with now."

Visit Cloud Wars for more.