Enterprise AI Economics: Context, Open Source, and Composite Roles
Arvind Jain discusses the shift from model hype to economic efficiency in enterprise AI. Key insights include the critical role of context in driving ROI, the acceleration of open-source adoption due to cost pressures, and the emergence of composite workforce roles. The analysis highlights how consumption pricing disrupts vendor bundling and why frontier models should be viewed as infrastructure assets.
The enterprise AI market is undergoing a critical inflection point where strategic focus is shifting from model capability to economic efficiency, operational control, and workforce restructuring. Arvind Jain, founder of Glean, argues that AI return on investment is fundamentally a throughput problem. Current deployments often suffer from brute-force context assembly, where agents waste excessive tokens searching for raw materials across disconnected systems. This inefficiency results in slow performance and inflated costs. To unlock true value, organizations must invest heavily in structured context layers that deliver precise, pre-assembled information to agents. By optimizing context delivery, enterprises can drastically reduce token burn, accelerate task completion, and transform AI from a cost center into a high-velocity productivity engine.
Open Source Economics and Geopolitical Friction
Rising inference costs are forcing a rapid migration toward open-source models. Jain predicts that the majority of enterprise workloads will run on open-source infrastructure within three years, driven by the need for cost control and data sovereignty. However, this transition is complicated by geopolitical risks. The emergence of high-performing models from Chinese providers has created a trust gap among Western enterprises. Despite technical parity and significant cost advantages, concerns regarding potential backdoors and regulatory exposure are slowing adoption, highlighting a tension between economic rationality and national security imperatives.
Consumption Pricing Neutralizes Bundling Threats
The competitive landscape is being reshaped by the shift from seat-based licensing to consumption-based pricing. While bundled suites like Microsoft Copilot leverage enterprise contracts to lock in users, consumption models undermine this strategy. When pricing is tied to actual usage, the economic advantage of bundling dissipates. Enterprises can now deploy best-of-breed AI tools that outperform bundled alternatives, paying only for the specific value generated. This dynamic empowers specialized vendors to compete on performance and efficiency rather than relying on vendor consolidation.
Workforce Evolution and Composite Roles
AI is catalyzing a fundamental change in role definition and team structure. Routine functions, such as data analysis and sourcing, are being automated, allowing business owners to query systems directly. Concurrently, the rise of composite roles is merging specialized functions—engineering, product, and design—into generalized positions. This consolidation requires smaller teams to deliver exponentially higher output. Jain warns that companies must do 10 times the work to get the same amount of revenue, emphasizing that headcount reduction must be paired with aggressive productivity gains to maintain competitive advantage.
Strategic Imperatives for Leaders
Success in the AI era requires viewing frontier models as infrastructure assets rather than competitors. Non-model companies should leverage the commoditization of the model layer to build differentiated application experiences. Leaders must also address the power law of token adoption, where a small fraction of users drive the majority of spend, by implementing governance and showcasing high-impact use cases. Ultimately, enterprises that master context architecture, optimize for open-source economics, and restructure around composite capabilities will capture disproportionate value in the coming years.
Key insights
-
AI ROI is determined by context efficiency rather than model selection alone. Brute-forcing context assembly burns tokens and slows agents, whereas structured context delivery reduces costs and accelerates throughput.
Impact: Organizations can lower AI spend by 30-50% while improving response times by investing in context architecture rather than chasing marginal model gains.
-
Open-source model adoption is accelerating due to cost pressures, with predictions of majority enterprise workload migration within three years. However, geopolitical concerns regarding Chinese providers create adoption friction.
Impact: Enterprises can achieve significant cost savings by adopting open-source models, but must navigate geopolitical risks and potential backdoor concerns through rigorous security audits.
-
Consumption-based pricing models erode the competitive advantage of vendor bundling strategies. When pricing aligns with usage, best-of-breed tools can compete effectively against bundled suites.
Impact: Specialized AI vendors can displace bundled offerings by demonstrating superior performance and efficiency, as enterprises prioritize value-based spending over seat consolidation.
-
AI is driving the emergence of composite roles that merge specialized functions, requiring smaller teams to deliver exponentially higher output. Routine data analysis roles are facing obsolescence.
Impact: Companies must restructure teams around generalist capabilities and implement rigorous productivity metrics to ensure that headcount reductions translate into tangible revenue growth.
-
Frontier model providers should be viewed as infrastructure assets by application-layer companies. Model commoditization expands the total addressable market and lowers barriers to entry for AI products.
Impact: Startups can focus resources on workflow integration and domain-specific value creation, leveraging the innovation and scale of frontier models to accelerate product development.
Action items
-
Audit current AI deployments for context inefficiencies. Identify workflows where agents are brute-forcing data assembly and implement structured context layers to pre-assemble information.
Impact: Reduces token consumption and latency, directly improving AI ROI and user experience without requiring model changes.
-
Evaluate open-source models for non-critical workloads to benchmark cost savings. Develop a risk assessment framework to address geopolitical concerns associated with specific model providers.
Impact: Enables cost optimization and budget predictability while maintaining security standards and regulatory compliance.
-
Redefine job descriptions to incorporate composite capabilities. Train employees to leverage AI for cross-functional tasks, merging engineering, product, and design responsibilities.
Impact: Increases organizational agility and output per employee, preparing the workforce for the transition to smaller, high-performance teams.
-
Implement token usage governance and showcase high-impact use cases in leadership meetings. Balance the power law of adoption by encouraging advanced usage without incentivizing waste.
Impact: Optimizes token spend distribution and drives cultural adoption of AI tools that deliver measurable business value.
Quotes
“Once you move towards consumption, there's no inherent bundling advantage.”
“You have to do 10 times the work to get the same amount of revenue from your customers.”
“For almost all other AI companies that are not doing frontier model training, they should see the model companies as a huge asset, not a competition.”