AI and Cloud Infrastructure Trends

Bessemer Venture Partners just released their State of the Cloud 2024 report. Here are the five top AI trends driving the most significant changes across technology and their predictions for what to expect by 2030. [1] Big Tech's new battle-of-the-century is for control of foundation models. • 90% of private GenAI funding came from corporate VCs, like Microsoft, Google, etc. 🔮 Prediction: The winners will reign supreme within cloud & compute markets. [2] AI is turning us into 10X developers. • Code copilots were widely adopted in 2023 and we're already seeing developments that suggest end-to-end automation could be arriving soon. 🔮 Prediction: Software developers will look more like software reviewers. Cost of development will fall. The average age of tech founders will fall. [3] Multimodal models (like voice, image, and video) and AI agents will transform human relationships with software. • Giving AI the equivalent of vision, hearing, and speech, opens the door for AI to impact work that requires these senses. 🔮 Prediction: In the next five years, Voice AI apps alone will unlock $10B of new software TAM. [4] Vertical AI could dwarf legacy vertical SaaS with new applications and business models. • Within every industry vertical, repetitive language tasks represent a significant share of activity. The Business & Professional Services industry is 10X the size of the software industry and is an example of this. 🔮 Prediction: Vertical AI's market cap will be at least 10X the size of legacy Vertical SaaS. The first Vertical AI IPO will occur in the next three years. [5] AI brings Consumer Cloud back from the dead. • Consumer cloud unicorns have historically been built after major tech shifts (i.g. mobile). Last month, OpenAI had more site visits than Reddit (2.5B vs 2.4B) 🔮 Prediction: By 2030 the top three businesses dominating the Attention Economy will be based on AI-generated content. Great takes from Lindsey Li and Bessemer team. Link to full report in comments.

So far this year, the U.S. data center landscape is shifting fast. As power constraints tighten in traditional hubs like Northern Virginia and Silicon Valley, secondary and emerging markets are stepping into the spotlight. Do you agree? Here’s what’s driving the shift and where the real opportunities may lie: 👉🏽 Key Trends: • Explosive Growth: U.S. data center inventory jumped 43% year-over-year in Q1 2025. Atlanta and Phoenix are now among the top four markets. • Hyperscaler Expansion: AI and cloud giants are pre-leasing space in non-traditional markets like Columbus, Des Moines, and Reno to secure power and speed to market. • Design Evolution: Liquid cooling and high-density racks are becoming the norm to support AI workloads. But what are the challenges? • Power Scarcity: Grid capacity is maxed in many metros. Transmission delays are pushing timelines into 2027 and beyond. • Construction Bottlenecks: Labor shortages, permitting delays, and rising material costs are slowing builds. • Cooling and Density Challenges: AI workloads demand new cooling strategies and facility designs, especially in retrofitted sites. So, where’s the opportunity? • Emerging Markets: States like Iowa, Indiana, and Ohio offer cheaper, more available power and faster permitting. This is ideal for hyperscaler and co-lo builds. • Behind-the-Meter Solutions: On-site renewables, gas peakers, and early-stage nuclear (SMRs) are gaining traction to bypass grid delays. • Sustainable Infrastructure: Clean energy integration is now a competitive edge. Why does this matter? AI infrastructure here in the U.S. is being built in places we’re only beginning to notice. Maybe even in a county near you! It’s a dynamic space, and I’m keeping a close eye on the trends and opportunities. What are you seeing in your region or sector? Let me know! 👇🏽

AI is driving a significant increase in cloud infrastructure spending, particularly compute and storage for cloud deployments. This trend leads to a shift towards high-capacity, GPU-heavy servers favored by hyperscalers, resulting in fewer shipped units. AI is the primary driver behind this surge, influencing the market dynamics and trends toward purpose-built hardware infrastructure tailored for AI-centric workloads. As a result, there may be challenges for rank-and-file IT leads in budgeting for AI capabilities, potentially neglecting traditional cloud services. Enterprises that have invested in cloud computing may face challenges as the market trends towards AI-centric infrastructure, leading to potential price fluctuations in equipment and cloud services. Strategic adaptation will be crucial for enterprises to optimize their investments and technology deployments amidst the evolving landscape of AI-driven cloud infrastructure. Check out my @InfoWorld blog on Friday, where I take this apart for you.

My latest in today's print edition of The Wall Street Journal: Many companies say the cloud is their go-to when it comes to training and running large AI applications—but today, only a small portion of existing cloud infrastructure is actually set up to support that. The rest is not. Now cloud providers, including Amazon Web Services (AWS) Services, Microsoft Azure and Google Cloud are under pressure to change that calculus to meet the computing demands of a major AI boom. “There’s a pretty big imbalance between demand and supply at the moment,” said Chetan Kapoor of product management at Amazon Web Services’ Elastic Compute Cloud division. Much of today’s cloud footprint consists of servers designed to run multiple workloads at the same time that leverage general-purpose CPU chips. A minority of it, according to analysts, runs on chips optimized for AI and servers designed to function in collaborative clusters to support bigger workloads. Now, cloud providers are working to make AI optimized infrastructure a greater proportion of their overall footprints. At the same time, other hardware providers may have an opportunity to make a play here. Dell Technologies expects that high cloud costs, linked to heavy use—including training models—could push some companies to consider on-premises deployments. “The existing economic models of primarily the public cloud environment weren’t really optimized for the kind of demand and activity level that we’re going to see as people move into these AI systems,” Dell’s Global Chief Technology Officer John Roese said. Read the full story here: https://lnkd.in/eMd7vv_H #ai #artificialintelligence #cloud #computing

🚀 Breaking: OpenAI + Oracle Supercharge the Stargate AI Data Center Buildout! Thrilled to share that OpenAI has just partnered with Oracle to add 4.5 GW of new AI-ready data center capacity in the US—taking Stargate’s total pipeline to 5 GW under development. Built to power over 2 million AI chips, this marks a major leap toward fulfilling the $500 billion / 10 GW AI infrastructure pledge made at the White House earlier this year. 🛠️ Why This Matters ⚡ Strategic Scale: This isn’t just about more land and servers—it’s adding the power and cooling infrastructure needed to sustain the next generation of AI innovation. 👷♂️ 100,000+ U.S. Jobs: From construction to skilled operations, this buildout is a massive economic win, spanning electricians, technicians, manufacturing, and service roles. 🇺🇸 Reindustrializing America: It reinforces AI infrastructure sovereignty—anchoring computation on U.S. soil with deep integration of compute, power, and supplier ecosystems. Stargate continues as a juggernaut—scalable, strategic, and fueled by partners like Oracle and SoftBank. Meanwhile, Microsoft remains a key cloud provider in the mix. 🔭 Perspective: This isn't just another data center deal—it's national-scale infrastructure in motion. For us in energy, grid, and data center sectors, collaboration and forward planning are more vital than ever. The trail already blazed by Stargate signals that AI-driven electrical demand and site readiness infrastructure need accelerated policy and planning support. What does this mean for your organization? Energy planners: Expect load growth that matches industrial rollouts. Grid operators: Time to align on transmission enhancements, microgrid solutions, and power resiliency. Developers & real estate teams: Optimize infrastructure/land bundling to meet AI hyperscale density. Policy leaders: Support permitting, workforce development, and strategic incentives for future-proof capacity. 🔁 Join the Conversation: How is your organization preparing for this new AI infrastructure wave? Are we aligned—or already falling behind in planning, permitting, grid-readiness? This is our moment to shape the future of power-forward AI.

Over the next 4 days, I’ll share my 4 biggest tech industry takeaways heading into 2024. The purpose of this exercise is to: 1. Unpack 2023’s biggest milestones (and explore what developers can learn from them) 2. Stay ahead of emerging trends in 2024 3. Predict which developer skills will be most in-demand in the coming months and years 🔮 Here’s takeaway #1: Cloud computing needs an overhaul ☁️ This year, we witnessed significant changes in cloud computing, driven by the rapid advancement of AI. Traditional cloud infrastructure, primarily based on CPU chips, is struggling to keep up with the demands of AI models. Companies like Microsoft, Amazon, and Google are rapidly adapting, transitioning to GPU-enabled servers and advanced AI processors. These developments have had huge ripple effects across the tech industry. But what does this mean for developers? AI is the new norm, reshaping best practices in cloud application development. Data management and security are becoming paramount, and we need developers with strong cloud skills to help drive this transformation. Here at Educative, we've seen this trend play out over the past year as many learners have taken advantage of our new immersive learning tool: Cloud Labs. With Cloud Labs, developers can get hands-on experience with AWS services — without any downloads, configuration, or cleanup. Though there are over 110 labs up on the platform today, here are a few of our most popular Cloud Labs in 2023: 🔹 Understanding Cloud Computing Essentials – Zero to Hero 🔹 Deploying a Machine Learning Model with Amazon SageMaker 🔹 Speech Synthesis Using Amazon Polly 🔹 Automating Data Processing with AWS Glue DataBrew If you are interested in adding cloud skills to your toolkit in the new year, Cloud Labs are a great place to start. >> https://educat.tv/41G6gBz #Cloud #AI #techcareer

July has been a pivotal month in global AI governance. Just days after the European Union released its final Code of Practice for general-purpose AI, the White House, on July 24, unveiled the AI Action Plan—a comprehensive national strategy to accelerate U.S. leadership in AI, focused on five key pillars. The AI Action Plan outlines a national approach to accelerating innovation, scaling infrastructure, and coordinating policy across sectors. It reflects a clear commitment to AI as a strategic priority—and sets the stage for long-term U.S. leadership in global AI development. The plan can be accessed on AI.gov - a recently introduced official U.S. government website to centralize and publicize the Trump Administration’s AI Action Plan and related initiatives: https://www.ai.gov/ * * * AI July 2025 Action Plan - Pillars: 1. Infrastructure: Laying the Foundation for AI The plan prioritizes large-scale, secure infrastructure to support AI development and deployment. Key initiatives include: - Accelerating construction of U.S.-based data centers, chip manufacturing sites, and compute clusters - Expanding energy capacity to power AI systems - Streamlining permitting to avoid infrastructure delays - Strengthening fiber-optic networks across regions 2. Innovation: Reducing Friction for AI Development To increase speed and scale, the plan focuses on: - Regulatory and procurement reform - Open-source support and incentives for smaller tech firms - Public-private innovation hubs and targeted R&D funding - Encouraging domestic model development and compute capacity 3. Workforce: Building AI Talent at Scale To meet demand, the plan supports: - National training programs and certifications - Partnerships with universities and community colleges - Accessible education and reskilling initiatives - Use of real-time labor data to match skills with needs 4. Global Leadership and Exports The plan positions U.S. AI capabilities as strategic assets by: - Promoting exports of the full U.S. AI stack - Advancing international cooperation and standards - Strengthening export controls to protect national interests 5. Federal Adoption and Public Sector Use Federal agencies are encouraged to lead by example through: - Wider deployment of AI tools across government services - Applications in education, healthcare, and national security - Clear procurement and safety guidelines An accompanying executive order ("Preventing Woke AI in the Federal Government, https://lnkd.in/g_vw3aJp) directs agencies to procure only AI systems that meet federal standards for accuracy, neutrality, and objectivity—reinforcing expectations for factual integrity in public-sector AI. * * * The plan has drawn strong industry support, with tech, energy, and manufacturing leaders welcoming its alignment on infrastructure, clear rules, and public-private partnership.

Just published with Lee Sustar: The Rise of #AI Cloud - https://lnkd.in/em7cmqRh "AI is your job (IT Ops) now too — another silo to bust. AI and cloud strategies can no longer live apart. It is imperative that cloud AI services and workloads are embedded in mainstream cloud operations. This means renewing the cloud operating model, adding in the AI capabilities required (i.e., activities, competencies, underlying technologies), fine-tuning operating unit structures, adjusting governance, and defining AI’s role in their broader cloud strategy and vision.