Data + AI Platform Providers

At FabCon and SQLCon in Barcelona, Microsoft made Fabric IQ generally available in Copilot Chat and Cowork, so business definitions from Power BI semantic models inform Copilot answers. It also previewed a Fabric data engineering agent, IQ sharing in OneLake, Fabric observability and Database Hub, and made BigQuery mirroring and DiskANN vector indexes generally available.

Buyer takeawayFabric customers can ground Copilot and agents in governed semantic models today, while most agent features remain in preview and need careful evaluation before production use.

Fujitsu reworked its Uvance business model around AI transformation, organised into consulting with forward deployed engineers, applications and AI agents, and AI platforms. It plans to release a new data and AI platform in March 2027 that combines its Kozuchi and Takane technology with frontier models and supports sovereign cloud. Fujitsu targets Uvance revenue above JPY 1.7 trillion by fiscal 2030.

Buyer takeawayFujitsu clients should expect a shift toward outcome-based and usage-based pricing and can plan sovereign data and AI workloads around the platform due in March 2027.

Oracle reported first quarter fiscal 2027 revenue of about $19.35 billion, up 30% year on year, with infrastructure revenue more than doubling to $7.4 billion and total cloud revenue up 62%. Remaining performance obligations reached $664 billion. The company guided to at least $90 billion in revenue for the full fiscal year and reported more than $30 billion in AI cloud bookings.

Buyer takeawayThe scale of committed AI cloud demand suggests Oracle will keep investing heavily in infrastructure and database services, though buyers should watch capacity allocation and pricing as large AI contracts absorb supply.

Snowflake reported second quarter fiscal 2027 product revenue of $1.49 billion, up 37% year on year, with total revenue of $1.55 billion and remaining performance obligations of $9 billion. Customers spending over $1 million annually rose 27% to 828. The company raised its full-year product revenue forecast to $6.07 billion and lifted its operating margin target.

Buyer takeawayAccelerating consumption suggests customers are expanding AI and data workloads on Snowflake, which supports its long-term viability as a strategic platform choice.

Snowflake and LSEG announced an expanded five-year enterprise collaboration at Snowflake World Tour London. LSEG increases its use of the Snowflake AI Data Cloud across its data, analytics and risk businesses, while Snowflake will license LSEG World-Check for its own customer onboarding. Joint customers can combine licensed LSEG content with their own data for governed analytics and AI.

Buyer takeawayFinancial services firms on Snowflake get easier governed access to LSEG content for AI and analytics work without building separate data pipelines.

Snowflake priced an upsized private placement of $3.75 billion in convertible senior notes, one day after proposing a $3.5 billion offering. The raise follows a quarter of faster product revenue growth and gives the company additional capital for investment in its AI data platform and other corporate purposes.

Buyer takeawayThe added capital gives Snowflake more room to fund AI product development and acquisitions, which reduces near-term vendor viability concerns for large customers.

Databricks bought Row Zero, a cloud spreadsheet startup founded by former AWS and Tableau engineers, for an undisclosed sum. It plans to combine the product with its Genie assistant so people and agents can work on governed enterprise data in a spreadsheet view without exporting files. Chief executive Ali Ghodsi said more acquisitions of this kind are planned.

Buyer takeawayEnterprises standardised on Databricks may be able to move spreadsheet-based analysis onto governed data, reducing uncontrolled data exports.

Databricks will invest more than US$350 million in Singapore over three years, growing local headcount from about 250 to over 500 and opening a larger regional headquarters. It named iFAST, Singapore Customs and Singtel as new customers in the region. The plan came alongside a partnership with Singapore's EDB and IMDA linked to the national AI strategy.

Buyer takeawayAsia Pacific buyers can expect deeper local support and delivery capacity from Databricks over the next three years.

SAP added the TabPFN-3.5 Plus tabular foundation model from Prior Labs, which it acquired in July 2026, to SAP AI Core. The model produces predictions directly from structured business tables without conventional training or tuning, and targets uses such as cash flow forecasting, payment delay prediction, supplier risk scoring and churn assessment. SAP has committed over one billion euros to Prior Labs.

Buyer takeawaySAP customers can test predictive use cases on ERP data with less data science effort, which may reduce reliance on separate machine learning platforms for routine forecasting.

IBM's September mid-month release for watsonx Orchestrate reorganised the product around build, AgentOps, governance and an AI Gateway in preview. It added content guardrails that can evaluate agent inputs and outputs with any model registered in a tenant, and a consolidated settings panel in the flow builder covering scheduling, data masking, translations and notifications.

Buyer takeawayTeams running agents on watsonx Orchestrate get finer control over content safety policies and a clearer path toward governing agents from a single console.