Michael Amori is CEO and cofounder of Virtualitics. A data scientist and entrepreneur with a background in finance and physics.
Over the last several months, the conversation about large language models (LLMs) has focused on capability: which models are most advanced, which proof points matter and which companies are moving too fast. The AI race has largely been framed as a race for better technology.
Those questions still matter, but a fundamental shift is underway within the enterprise as AI capabilities become increasingly accessible through cloud providers, software platforms and enterprise ecosystems. Much of the dialogue treats all AI use cases the same when they can be fundamentally different.
AI, in general, is becoming abundant and increasingly accessible, but AI with context remains scarce.
For most organizations, the challenge has shifted from obtaining intelligence to applying it effectively, considering an organization’s unique data, workflows, policies, governance and institutional knowledge.
As AI becomes more widely available, intelligence will become less differentiated, and competitive advantage will be derived from the context a model operates within. The ability to transform intelligence into trusted decisions and measurable outcomes will define the next era of enterprise AI.
Why Context Matters
Enterprise decisions rarely depend on information alone. Operational priorities, regulatory requirements, risk tolerance and years of accumulated expertise shape them. AI can surface insights and generate recommendations, but without context, it may struggle to know which factors matter most.
Consider a supply chain leader evaluating a potential disruption. An AI model may identify the issue and summarize the data, but the decision requires context. Which program is most critical? Which trade-offs align with business objectives? The model provides intelligence, but context determines how to turn insight into action.
At Virtualitics, we see this in complex operational environments where decisions depend on connecting maintenance history, parts availability, scheduling constraints and organizational priorities. AI agents help surface trade-offs and explain recommendations while keeping final judgment with the people responsible, made possible by the data, knowledge, governance and expertise surrounding the model.
How To Achieve Context Advantage
Organizations building sustainable AI advantage invest in capabilities well beyond the model. Four elements define what I believe will become a context advantage:
1. Proprietary Data
The most valuable enterprise data often sits inside the firewall, where LLMs alone can’t reach it. Operational records, maintenance histories, customer interactions and domain-specific datasets give organizations perspectives that generic models don’t have.
As AI capabilities become a level playing field, unique data becomes increasingly valuable, often determining the quality of AI outcomes.
2. Institutional Knowledge
Every organization holds expertise that exists beyond formal documentation. Experienced operators understand nuances and decision criteria that no database can capture. They see patterns others miss and understand how decisions affect broader outcomes.
Capturing this knowledge may be one of the biggest opportunities in enterprise AI. Combined with institutional expertise, AI moves organizations beyond automation and toward organizational intelligence.
3. Governance And Operational Frameworks
Trust remains foundational to enterprise AI. Leaders need confidence that recommendations align with priorities, regulations and policy, with transparency into how they were generated.
Context includes the governance that guides how AI operates and how outputs are evaluated. Without it, intelligence can scale faster than accountability.
4. Strategic Partnerships
No organization has every source of expertise, data or technology needed to maximize AI outcomes, which is why strategic partnerships are becoming critical to context advantage. They connect LLM capability with the domain expertise and operational realities to make it useful.
For example, we recently announced a strategic collaboration with OpenAI to bring frontier models into Iris, its agentic AI for readiness. The goal is to combine advanced reasoning with the readiness, personnel, material and sustainment context behind mission-critical decisions.
The same logic applies across our broader ecosystem. Partners in cloud, technology, integration and services help customers connect complex data with AI, implementation support and specialized expertise, expanding the context available and increasing the odds that insight becomes action.
In the next phase of enterprise AI, the advantage will belong to those who assemble the richest ecosystem of data, expertise and partnerships around it.
An Executive Playbook For Building Context Advantage
As leaders evaluate AI investments, five questions are worth asking:
• What unique context exists within our organization that competitors can’t easily replicate?
• How effectively can AI access trusted enterprise data?
• Are we capturing institutional knowledge before it becomes siloed or lost?
• Which partnerships strengthen our AI ecosystem and improve decision quality?
• How do we ensure AI recommendations reflect our priorities, governance and business objectives?
These questions move the discussion beyond model selection toward long-term value creation.
The Next AI Race
The first phase of enterprise AI was about access, as organizations raced to adopt and deploy new models. The next phase will look different.
As AI becomes more abundant, advantage will increasingly come from the context organizations build around it. Data, expertise, governance, workflows and partnerships become the ingredients that turn AI outputs into trusted decisions and measurable outcomes.
Models will keep improving, and access will keep expanding. However, the lasting advantage will belong to the organizations that build the richest context around the most advanced AI.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?