Monthly Trends
3 conversational turns exploring this topic
Q: What rose in importance this month in enterprise AI?
Subject: Ins7ghts: The REAL Enterprise AI Story This Month (It's Not What You Think)
Hey Data Pros,
Forget the hype about shiny new models. This month's surge in AI isn't about algorithms; it's about infrastructure and geopolitics. The data doesn't lie:
| Entity | Category | Growth (%) | Key Takeaway |
|---|---|---|---|
| Microsoft | Organization | 13,800% | The Azure/OpenAI partnership is solidifying its dominance in the enterprise. This isn't just about access to models; it's about the full stack: compute, data storage, and AI services deeply integrated. |
| OpenAI | Organization | 24,100% | Obvious, but the sheer magnitude is staggering. They're not just a model provider; they're becoming the de facto AI platform for many enterprises. |
| U.S. | Location | 5,360% | Increased focus on AI regulation, national security implications, and investment in domestic AI capabilities. This will directly impact how enterprises deploy AI. |
| Azure | Organization | 8,300% | Underscores Microsoft's strategic play. Azure is becoming the platform for enterprise AI, leveraging OpenAI's models. |
| China | Location | 8,250% | The US isn't the only one thinking about geopolitics. Expect increased competition and potentially bifurcated AI ecosystems. |
| Claude | Person | 33,100% | Anthropic's Claude is gaining serious traction, especially among enterprises seeking alternatives to OpenAI. The focus on safety and explainability is resonating. |
| Meta | Organization | 8,025% | Their open-source approach with Llama is a calculated move to democratize AI and challenge the closed ecosystems of Microsoft/OpenAI. Expect more enterprises to experiment with Llama. |
My Hot Take: The real battle isn't about who has the best model today. It's about who controls the infrastructure and ecosystem that enterprises will rely on for the next decade. Microsoft is playing chess while others are playing checkers.
The Claude Bump: Don't sleep on Anthropic. While OpenAI gets all the headlines, Claude's focus on safety and constitutional AI is a major differentiator, particularly for regulated industries. The rise of "Claude" in the data reflects a growing enterprise demand for responsible AI.
The Geopolitical Squeeze: The rise of both the U.S. and China in the data is a flashing warning sign. Enterprises need to start thinking about data sovereignty, compliance with different regulatory regimes, and the potential for AI technologies to become weaponized.
Advice for Data Professionals:
- Deep Dive on Azure AI: If you're in the enterprise, understand the Azure AI stack inside and out. It's becoming table stakes.
- Explore Claude: Don't get locked into the OpenAI ecosystem. Evaluate Claude for use cases where safety and explainability are paramount.
- Think Globally, Act Locally: Understand the AI regulations in the regions where your company operates. Data residency and compliance are no longer optional.
- Experiment with Llama: Meta's open-source approach offers flexibility and control. Explore Llama for internal projects and custom AI solutions.
This isn't just about building cool AI applications. It's about building a sustainable and responsible AI strategy that aligns with the long-term interests of your organization.
Stay sharp,
Yves Mulkers
Data Strategist | Ins7ghts Newsletter
Q: Why is the top one rising?
Subject: Ins7ghts: The REAL Enterprise AI Story This Month (It's Not What You Think) - Part 2
Hey Data Pros,
Forget the hype about shiny new models. This month's surge in AI isn't about algorithms; it's about enterprise readiness. And the biggest signal of that is the rise of Microsoft 365 Copilot.
You asked why? It's simple: it's not about if you can do AI, it's about how you deploy it at scale, securely, within the workflows your people already use.
Let's be honest, everyone's been drooling over the open-source LLMs and building their own RAG pipelines. That's cool for a hackathon, but try getting that past your CISO.
Microsoft is betting (correctly, I think) that enterprises will prioritize integrated, secure, and compliant AI, even if it's not the absolute bleeding edge.
Here's the breakdown:
| Driver | Why It Matters |
|---|---|
| Productivity Needs | The C-suite gets this. They understand ROI. AI that boosts output is a no-brainer (in theory). |
| Microsoft's Strategic Focus | MSFT isn't just throwing AI at the wall; they're weaving it into their entire ecosystem. Lock-in, baby! |
| Security & Compliance | This is the real battleground. Enterprises need to trust their AI, and MSFT is selling that trust. |
| Seamless Workflow Integration | No one wants to learn a new tool. Copilot lives where people already work. |
My hot take? The real story isn't Copilot's capabilities (which are still evolving), but its distribution. Microsoft has a massive, captive audience. They're turning AI into a feature, not a project. This is the "iPhone moment" for enterprise AI. It's not the most powerful device, but it's the one everyone will use.
Contrast this with the open-source crowd. While innovation is happening there, the path to enterprise adoption is still murky. Who's going to provide the support, the security audits, the compliance guarantees? It's a fragmented landscape, and that's a huge barrier.
Here's my advice:
- Stop chasing the shiniest new model. Focus on practical applications within your existing infrastructure.
- Evaluate Copilot (or similar integrated solutions) seriously. Don't dismiss it as "just Microsoft." Understand the security and compliance benefits.
- Demand proof of ROI. Productivity gains are great, but quantify them. Track usage, measure output, and demonstrate the value of AI to your stakeholders.
The future of enterprise AI isn't about raw power; it's about accessibility, security, and integration. Microsoft is betting on that, and so should you.
Until next time,
Yves.
Q: What should my client know?
Subject: Re: Ins7ghts: The REAL Enterprise AI Story This Month (It's Not What You Think)
You asked why data infrastructure is rising in importance. Simple: everyone's realizing that even the shiniest new model is useless if you can't actually use it.
Your client needs to understand this: AI isn't about the algorithm, it's about the plumbing. Think of it like this: everyone's obsessed with the faucet (the AI model), but nobody's thinking about the pipes bringing the water in (the data infrastructure).
Here's what I'm seeing based on this week's data:
Money Talks: Anthropic just secured another $10 BILLION+ in funding. But read between the lines: that money isn't just for model development. It's for building out the infrastructure* to support those models at scale. They know, and the smart money knows, that's where the real bottleneck is.
The Regulatory Squeeze: Expect 2026 to be a battleground between federal deregulation and state-level AI rules. This means compliance and governance are about to get much more complex. Your client needs a data infrastructure that can handle that complexity now*.
- The Attention Problem: Even at the cutting edge, researchers are wrestling with fundamental problems like "oversmoothing" in attention mechanisms (see DyDiLA). This highlights the need for robust data pipelines that can handle the nuances of these models and prevent garbage-in, garbage-out scenarios.
The Pain Points Are Real:
| Pain Point | Why It Matters