AI platform providers are racing to incorporate tools that strengthen agentic AI governance, improve visibility into and control over token usage, and support a range of deployment options. To capture the hype around agentic AI, several AI platform providers have rebranded their solutions over the past year.  Their platforms now lead with agentic AI, although they also encompass tools to build predictive machine learning models and support generative AI models. Within this rapidly evolving landscape, Google and Amazon stand out as clear leaders in cloud-based enterprise AI platforms, according to GlobalData, a leading intelligence and productivity platform.

GlobalData’s latest report, “Cloud-based Enterprise AI Platforms: Competitive Landscape Assessment,” reveals that sovereignty concerns are top of mind as enterprises seek to have greater control over critical operations.  Similarly, guardrails that include the ability to disable agents are in demand, in addition to improved observability tools. 

Google and Amazon offer a full suite of model development and management tools, as well as access to leading large and specialized language models, and complementary data storage and management solutions that create a more seamless experience for end users.

Rena Bhattacharyya, Chief Analyst and Practice Lead, Enterprise Technology and Services at GlobalData, comments: “The platforms have expanded to include capabilities that help organizations incorporate, customize, and evaluate large language models (LLMs), as well as deploy AI agents. They are increasingly offering features that help organizations govern agentic AI and deploy responsible and ethical AI, such as the ability to create audit trails and track AI agents built on other platforms.”

GlobalData notes that, to address increasing concern over the runaway costs of AI inference, platform providers are also rolling out tools to better manage token usage.  The tools allow users to set limits based a range of criteria, including by person, department, or project.

Alfie Amir, Senior Principal Analyst, Enterprise Technology and Services at GlobalData observes: “Frustrations over measuring ROI and meeting project expectations are spreading. Enterprises are moving from a strategy of ‘token maxxing’ to one of ‘value maxxing’ to focus more on the success of a project than on the amount of AI consumption.” 

Bhattacharyya concludes: “Google’s internally developed suite of language models is broad and impressive.  It was early to market with AI agent development tools, and it demonstrates thought leadership by developing the Agent2Agent (A2A) protocol with its peers. Amazon boasts a broad portfolio of solutions, from data management to AI agent development, as well as strong partnerships with IT services providers and a marketplace that provides access to partners’ solutions.”