- According to the report, over 70% of active cloud computing environments utilize managed AI services, including Azure OpenAI Service, Amazon SageMaker, and Google Cloud’s Vertex AI.
- A possible explanation for this caution is that enterprises are still wary of the novel security challenges brought about by generative AI.
Wiz Inc.’s recent research underscores the rapid growth in the utilization of managed artificial intelligence services and self-hosted AI tools within cloud environments.
The remarkable expansion of AI is propelled by the emergence of generative AI models like ChatGPT, elevating the technology to popularity comparable to the widely used open-source Kubernetes software, a core component in numerous modern business applications. Released recently, Wiz’s State of AI in the Cloud 2024 report draws from an extensive sample size exceeding 150,000 public cloud accounts. This report presents compelling evidence of enterprises’ fervent desire for AI tools to improve business processes and lower operational costs.
According to Wiz’s researchers, generative AI has swiftly risen in the last 18 months to claim the position of the most vibrant segment in today’s technology industry. While traditional AI and machine learning services have existed for several years, integrated into numerous products, the ascent of generative AI, fueled by large language models, has propelled the technology into the forefront of public awareness.
The ascent of generative AI began in July 2022, marked by the beta releases of image generation models like OpenAI’s DALL-E 2 and Midjourney. The momentum further accelerated with the introduction of ChatGPT toward the conclusion of that year.
ChatGPT captured the public’s imagination like no other AI model had before, sparking a frenzied interest among businesses eager to explore how to integrate this technology into their products and services. Wiz reports that this surge in interest has fueled the growth of a burgeoning community of AI builders, leading to the development of services and tools dedicated to the training, fine-tuning, management, and deployment of AI models.
Artificial Intelligence Takes Control of the Public Cloud
Wiz highlights that generative AI is inherently cloud-native due to its highly compute-intensive training and inference processes. This necessitates organizations to leverage the unparalleled scalability of cloud computing infrastructure.
According to the report, over 70% of active cloud computing environments utilize managed AI services, including Azure OpenAI Service, Amazon SageMaker, and Google Cloud’s Vertex AI.
This indicates a remarkable rate of adoption. In comparison, Wiz’s research reveals that the widely used Kubernetes container orchestration software, fundamental to the majority of present-day business applications, has been implemented in 81% of all cloud environments.
Thus, despite being a relatively new concept, AI services are already nearly as common as one of the most popular and extensively utilized open-source software projects ever created. Wiz’s researchers said, “To us, this indicates that cloud-based AI services are seeing an unusually high rate of adoption relative to other services.”
Considering Microsoft Corp.’s strong affiliation with OpenAI, it is not surprising that Azure AI Services has risen as the leading fully managed AI service in the cloud. Wiz reports a remarkable 228% surge in new users for Azure OpenAI in a mere four-month span in 2023.
However, Microsoft’s competitors are in close pursuit. While 54% of all Azure environments incorporate managed Azure OpenAI instances, Amazon’s counterpart, SageMaker service, is present in 53% of AWS environments. Vertex AI or AI Platform instances are found in 44% of Google Cloud environments.
Most Implementations are Experimental Due to Ongoing Security Concerns
Wiz asserts that while the expansion of AI services is nearly unmatched, 32% of enterprises are currently in the experimentation phase with the technology, deploying fewer than ten instances of AI services in their cloud environments. According to Wiz, this suggests that a significant number of cloud customers are not fully prepared for large-scale AI deployment.
A possible explanation for this caution is that enterprises are still wary of the novel security challenges brought about by generative AI. Similar to the initial phase of cloud computing, many new AI services are being implemented without well-defined standards and governance, posing significant risks for enterprises.
Legitimate concerns arise due to the well-documented potential of AI to propagate misinformation and harmful content in the past year. Furthermore, there are challenges related to data security, exemplified by Microsoft’s recent inadvertent exposure of 38 terabytes of AI data.
2024 is Set to Be a Crucial and Defining Year for Generative AI
According to Wiz, this year will witness enterprises intensify their explorations into generative AI. The cost associated with AI model training and inference, along with the imperative for heightened security in AI usage, will emerge as crucial factors for businesses moving forward—assuming they haven’t already—given the seemingly unstoppable trajectory of this trend at present.
Certainly, 2024 is poised to be a pivotal year for technology, with numerous companies expected to make decisions on worthwhile use cases for investment and determine the types of products and features they plan to develop.