The New Cloud Wars: The Profound Impact of AI on Cloud Consumption

The cloud has been the cornerstone of our digital infrastructure for several years, reshaping industries from healthcare to retail to finance. In pandemic years alone, thousands of organisations have integrated cloud services into their core operations, fine-tuning and scaling them as demands grew.

Just as consumption patterns seemed to stabilise, a disruptive force has arrived: artificial intelligence (AI). Generative AI, in particular, is not just another trend but a force fundamentally reshaping cloud dynamics in ways even the most seasoned cloud users didn’t see coming.

As AI tools mature and adoption accelerates, cloud costs and energy demands are soaring. This new age of AI doesn’t make cloud computing unsustainable. However, it does push cloud providers and users to rethink their strategies critically. It’s not just about managing costs; it’s about managing them strategically.

Let’s delve a little deeper.

Crests and Troughs in Cloud Consumption

Cloud adoption has been a rollercoaster, experiencing substantial flux over the past several years. Initially, the transition to the cloud premise was the promise of agility and scalability. The pandemic triggered an unprecedented spike in cloud usage as organisations scrambled to support remote operations and digital transformation.

It seemed like the cloud was at its peak. Companies poured into cloud ecosystems, leading to storage and compute usage surges.

However, costs began to swell in parallel. Many businesses underestimated the cloud’s long-term costs, and before long, CFOs searched for ways to rein in spending. Cost management tools emerged but often fell short in accuracy and efficacy.

Traditional budgeting practices proved inadequate as many organisations faced complex multi-cloud and hybrid environments, each with its cost structures.

Then, there’s the oligopoly at the heart of cloud infrastructure. AWS, Google Cloud, and Microsoft Azure are the titans in this space, leaving little room for competition. The few existing alternatives do not offer a compelling advantage, or their presence is limited.

Despite these hurdles, public cloud infrastructure remains the most viable choice for businesses of all sizes. It enables enterprises to access computing power on demand without massive capital expenditure, a capability that no on-prem solution could offer with equal flexibility or cost-effectiveness.

Today, the cloud is evolving yet again. As AI models, especially those powering generative AI, surge in popularity, cloud consumption patterns are shifting, and with them, the economics of cloud utilisation.

Gen AI is a Power Guzzler

In the generative AI era, power is king. Or, more accurately, electricity is king.

Generative AI systems consume vast energy due to the heavy computing requirements for training and deploying large language models and other AI algorithms. Data centres worldwide are feeling the strain — Morgan Stanley found that energy demand from generative AI will grow by 70% annually through 2027.

Therefore, the demand for cloud computing resources is skyrocketing. AI workloads are data — and compute-intensive, necessitating immense storage, memory, and GPU resources. Generative AI, in particular, relies on massive datasets and high-performance processing.

This dependency naturally pushes cloud providers to invest in additional infrastructure, including specialised AI hardware like GPUs and TPUs. In turn, the costs of operating these resources are passed on to users through a price increase.

Remember, such demand isn’t limited to private clouds either; public cloud providers are also expanding data centres to accommodate this new generation of AI applications. But there’s a catch: building and maintaining energy-intensive data centres don’t come cheap.

Rising operational costs could push providers to raise prices or seek sustainable energy sources to mitigate expenses.

Ultimately, this convergence of AI demand and cloud processing growth forces cloud users to grapple with the complex economics of balancing compute resources and cost.

5 Ways AI is Reshaping Cloud Consumption

Generative AI is rewriting the rulebook for cloud usage, with far-reaching implications for everyone, from small startups to Fortune 500 companies. Here are five ways it’s driving change:

1. Generative AI is intensely data-reliant

The hunger for data has reached new heights as generative AI models require enormous datasets for training. Each data point, each learning cycle, translates to compute cycles and, in turn, cloud consumption. Organisations have to budget for this massive increase in data storage and processing.

2. Gen AI-based SaaS applications are soaring in demand

Software-as-a-service (SaaS) applications increasingly embed AI capabilities, driving demand for cloud resources. From AI-driven marketing tools to content creation platforms, SaaS vendors are pushing workloads to the cloud, further compounding demand.

3. The citizen-coding and startup boom

Generative AI has empowered citizen coders and entrepreneurs to bring new apps to market without extensive programming experience. With just a few clicks, a non-coder can build an app on cloud platforms that leverages AI for unique use cases. This democratisation is spurring an influx of new cloud-dependent apps.

4. Existing cloud applications are incorporating Gen AI

Companies retrofit existing applications with generative AI features to enhance users’ functionality. It has led to a rise in hybrid workloads that blend traditional application models with AI, requiring more cloud resources for seamless operations.

5. Cloud-native AI app building made more accessible

Building and deploying AI-powered applications has always been challenging. Cloud-native tools enable developers to integrate generative AI directly into their products, creating a seamless bridge between innovation and scale. As a result, more applications are being built from the ground up with AI, making cloud consumption an integral component.

My Closing Thoughts: How to Avoid Overselling AI Benefits While Staying Bullish on the Cloud

Generative AI has brought a wave of optimism — and with good reason. The promise of more innovative tools, efficient workflows, and endless innovation is hard to resist. But it’s easy to get swept up in the hype and overlook the challenges lurking beneath. For many, managing the rising costs and energy demands of AI-driven cloud consumption is becoming increasingly complex.

Cloud users and providers alike must ask the question: How can we balance our AI ambitions with practical, sustainable cloud strategies? By doing so, we can ensure that we are not just embracing AI but harnessing its power in a way that benefits our digital ecosystem.

The key lies in tempering enthusiasm with clear-eyed realism. AI offers tremendous potential, but not every workload needs to go on the cloud, and not every cloud model will fit AI’s needs.

To avoid overselling AI’s cloud transformation, focus on defining clear usage guidelines, robust cost management practices, and sustainability measures. By doing so, you can ensure that AI’s cloud transformation is not a strain but a well-managed evolution of your digital ecosystem.

If you want to continue this conversation, please email me at Arvind@AM-PMAssociates.com.