AI Platforms Becoming Strategic to Business — ISG - June 24, 2025

As AI adoption and deployment surge, enterprises need to select the right AI platform to maximize the value of their people, processes and data, according to new research from global AI-centered technology research and advisory firm Information Services Group (ISG).

The ISG Buyers Guides for AI Platforms, produced by ISG Software Research, provide the rankings and ratings of over 30 software providers and their products to support assessment and selection for enterprise use of AI. The research finds that the growing awareness of generative AI in recent years has intensified companies’ focus on developing, deploying and maintaining all kinds of AI applications. Organizations are increasingly adopting AI platforms to translate AI concepts into opportunities for increased innovation and productivity.

“Fewer than one in five enterprises is at the highest level of innovation in using analytics and data,” said David Menninger, executive director, ISG Software Research. “AI capabilities — along with improved skills and processes — let companies tap into rapid, ongoing advancements in this area.”

David Menninger, at ISG

AI has existed for many decades, but its adoption and use have grown significantly since scale-out computing and object storage made it economically feasible for more enterprises to collect and process massive amounts of data, the research finds. More recently, GenAI has made AI more accessible and easier for employees and the public to use. Enterprises are on board for this transition: 85 percent of companies consider investment in GenAI important or critical in the next 24 months, ISG says.

Comprehensive AI platforms help enterprises access and prepare data for AI models, train and test those models, and then deploy and maintain them, the report says. In all phases of development and operation, qualified AI platforms ensure all models comply with internal policies and regulatory requirements, including retraining and redeployment as needed to bring them back into compliance. Platform providers will be important to many AI initiatives as companies race to adopt the technology. Through 2027, more than half of enterprises will realize they lack sufficient AI skills and will make new investments to avoid being surpassed by rivals, ISG has found.

While GenAI and agentic AI adoption grow, many organizations will also use AI platforms to support traditional and predictive AI where these remain most powerful, the report says. For example, predictive AI is expected to have a bigger impact than GenAI on applications involving credit risk, fraud detection, algorithmic trading and customer acquisition.

AI platform providers are ahead of the enterprise demand for software to help them move AI to production, keep models up to date and maintain governance, ISG says. For these functions, called machine learning operations (MLOps) or large language model operations (LLMOps), enterprises until recently had to create scripts and cobble together solutions. Now both general and specialist vendors offer tools, and ISG believes that four out of five companies will use MLOps and LLMOps tools by 2027.

ISG advises enterprises to approach AI adoption methodically. Choosing the right provider and product is critical, but so is improving skills and processes. To be as competitive as possible, companies should make a balanced set of upgrades to both technology and the organization.

For its 2025 Buyers Guides™ for AI Platforms, ISG evaluated software providers across three platform categories — AI Platforms, Agentic and Generative AI and Machine Learning and Large Language Model Operations — and produced a separate Buyers Guide for each. A total of 35 providers were assessed: Alibaba Cloud, Altair, Alteryx, Anaconda, Anthropic, Automation Anywhere, AWS, C3 AI, Cloudera, Cohere, Databricks, Dataiku, DataRobot, Domino Data Lab, Google Cloud, H2O.ai, Huawei Cloud, Hugging Face, IBM, MathWorks, Microsoft, NVIDIA, OpenAI, Oracle, Palantir, Quantexa, Red Hat, Salesforce, SAP, SAS, ServiceNow, Snowflake, Teradata, UiPath and Weights & Biases.

ISG Software Research rates software providers in seven evaluation categories. Five are related to product experience related: usability, manageability, reliability, capability and adaptability. Two are related to customer experience: validation and total cost of ownership and return on investment (TCO/ROI). Providers ranked in the top three for each evaluation category are named as Leaders. Within each platform category, those with the most Leader rankings are named as Overall Leaders.

The Overall Leaders of the 2025 Buyers Guides for AI Platforms were the following:

AI Platforms: Oracle scored the highest overall rating, with Google Cloud in second place, followed closely by IBM. Oracle was designated a Leader in six categories, Google Cloud in three and IBM in one. In addition to the top three providers, Alteryx, AWS, Cloudera, Databricks, Dataiku, Domino Data Lab, Microsoft, Salesforce, SAP, SAS, Snowflake and Teradata also were rated Exemplary. Alibaba Cloud, Altair, DataRobot, H2O.ai, Hugging Face, MathWorks and Palantir were rated Innovative.

Agentic and Generative AI: Google Cloud topped the list, followed by Oracle and IBM. Google Cloud was designated a Leader in four categories, Oracle in six and IBM in one. In addition to these providers, Alibaba Cloud, Automation Anywhere, AWS, Databricks, Dataiku, DataRobot, Domino Data Lab, Microsoft, NVIDIA, Salesforce, SAP, ServiceNow, Snowflake, Teradata and UiPath also were rated Exemplary. H2O.ai, Hugging Face, OpenAI, and Palantir were rated Innovative.

Machine Learning and Large Language Model Operations: Oracle scored the highest rating, followed by AWS and Databricks. Oracle was named a Leader in all seven categories. AWS was named a Leader in two categories and Databricks in four. In addition to these providers, Alteryx, Cloudera, Dataiku, Domino Data Lab, Google Cloud, IBM, Microsoft, SAS, Snowflake, and Teradata were rated Exemplary. Alibaba Cloud, DataRobot, H2O.ai, and MathWorks were rated Innovative.

“Not all AI software is the same. Nuanced differences can determine the fate of its use across the enterprise for traditional, agentic, and GenAI, along with MLOps and LLMOps,” said Mark Smith, partner and chief software analyst, ISG Software Research. “This research gives enterprises a holistic view of each provider’s AI products that will help them make critical selection decisions.”

To learn more, visit: www.isg-one.com