Everpure leveraged its Pure Accelerate 2026 media briefing in Singapore to narrate that storage is no longer a commodity and data – not just apps or models alone – is the strategic lever for successful AI initiatives.
At the centre of its data management strategy is 1touch, the data intelligence company Everpure acquired this May and whose technology Everpure says will help customers discover data across environments, attach business context and apply governance controls before data is used by AI systems.
From application silos to data context
Nathan Hall, Everpure’s vice president and general manager for APJ opened the briefing by pointing to the company’s recent commercial momentum and its effort to grow recurring services such as Evergreen One alongside larger strategic infrastructure engagements.
But the briefing’s larger theme was architectural. Par Botes, Everpure’s vice president of AI infrastructure, argued that enterprise technology has historically been built around applications, with data organised according to each application’s own logic and terminology.
“What happened over time is each application created its own semantics,” Par said. “The finance application thought of a buy order in one way. The Salesforce application thought of this order in a different way. The support application thought about this order in a third way. The supply chain application that shipped the equipment to the customer thought about (the same order) in (yet) another way.”
This often forces enterprises to build translations between systems – sometimes through spreadsheets – before they can produce a unified view of a customer, order or operational process.
Everpure’s proposed alternative is to make data, rather than individual applications, the enduring system of reference. In that model, applications become consumers of governed data, while metadata, business meaning and policies are managed independently of the application systems that first created the information.
1touch.io and the data readiness challenge
The strategy helps explain Everpure’s move to acquire 1touch. The company positioned 1touch as a way to add data discovery, classification, and contextual intelligence to its platform.
Ashish Gupta, previously 1touch.io’s chief executive and now general manager of Everpure’s data management business, said enterprises first need visibility into what data they have and where it resides.
“We were acquired to help our customers really understand their data,” Ashish said. “Doing this requires you to discover the data regardless of where it sits, whether it’s structured, unstructured, on-premises, in the cloud, on Everpure or other storage, and then classifying it based on sensitivity.”
1touch.io’s capabilities originally developed for data privacy and compliance like GDPR are increasingly necessary for AI to work. AI agents may pull information from multiple enterprise sources, making it important to know what data was retrieved, whether it was current and authorised, and which policies should govern its use.
Because AI agents consume data from across the enterprise, they need that underlying context, meaning, and governance attached to the data to avoid errors and maintain compliance.
We’re building auditability in everything we do.Par Botes
Par also described his vision of a future where data is no longer locked into specific applications. Instead, data carries its own built-in context, meaning, and capabilities so that any application across the enterprise can consume it on demand.
AI needs auditable data pipelines
Everpure also cited an Omdia study in making the case that organisations remain unprepared for enterprise AI. According to figures presented by the company, 97-percent of organisations are struggling to transition from AI pilot to production, 68-percent of IT leaders rank data management as the top challenge when moving AI into production. and 58-percent lack the data environment visibility required to support successful AI initiatives.
The underlying issue is familiar to enterprise technology leaders – valuable data is often fragmented across cloud services, on-premises systems, SaaS applications, mainframes and unstructured repositories, with inconsistent ownership and uneven governance.
Par and Ashish framed data discovery, classification and contextualisation as infrastructure requirements rather than discrete compliance activities.
Ashish described, “You need to ensure that you can discover data regardless of where it sits. It needs to be discovered and then it can’t be just classified by a tag. You need to be able to open that file, understand what’s the data inside that file… and then contextualise it.”
Par also sees context, traceability and key data management objectives as core infrastructure concerns. “We’re building auditability in everything we do. For every single query you do, we track the query. You can go back and look at it after the fact. You can figure out exactly why the answer was the way it was, which data supported making that answer.”
You need to ensure that you can discover data regardless of where it sits. It needs to be discovered and then it can’t be just classified by a tag. You need to be able to open that file, understand what’s the data inside that file… and then contextualise it.Ashish Gupta
The combined objective is to make information explainable and governable across the enterprise without losing the business meaning and policy controls associated with it.
Proof points
Nathan also shared that the company had secured a design win and a supply agreement with a second top-five hyperscaler, describing it as a validation of its technology at hyperscaler environments.
He said Everpure was the only storage company supplying technology at the hyperscaler backbone level.
Everpure has pointed to analyst recognition as evidence of its momentum. In a separate announcement ahead of its FY27 second quarter earnings call, the company said Gartner had positioned it highest for execution and furthest for vision in the 2026 Magic Quadrant for Enterprise Storage Platforms.
Explainable AI
For AI and data leaders, Everpure’s message is straightforward: infrastructure decisions for AI cannot end with compute capacity, or model selection.
AI systems need governed access to data that has been located, classified, understood and monitored over time. Without that foundation, enterprises can struggle to establish that an AI-generated answer was based on authorised, relevant and current information.