2026 State of AI: Bi-Annual Snapshot
The Execution Era of AI
1. Differentiation Has Moved to the Application Layer
2. Multi-Model Architectures Are Becoming the Standard, Driven by Cost and Control
3. Signals of AI Monetization Evolving, But No Final Answer
4. AI Is Acting as a Force Multiplier Across Organizations
5. The AI Tooling Ecosystem Is Maturing
Over the past six months, the AI market has shown signs of entering a new phase of maturity.
What started as the race to experiment with large models and launch early AI features has increasingly evolved into a challenge of scaling AI into durable, economically sound products.
Drawing on survey data from ~300 executives building AI products, alongside deep operating insights from the ICONIQ Community, our bi-annual State of AI report captures how companies building AI products are navigating this transition, from model strategy and product differentiation to agentic workflows and monetization.
Across the data, in our view, one pattern is unmistakable. While AI is now embedded in many product roadmaps and organizations, outcomes are diverging. The advantage is accruing to teams that can scale AI into production reliably, manage costs thoughtfully, and integrate it deeply into workflows that matter.
The takeaway to us is clear: AI leadership in 2026 will be defined by disciplined execution across product, cost, trust, and go-to-market.
1. Differentiation Has Moved to the Application Layer
A strong signal is the decisive shift of our respondents away from model-level differentiation.
~70% of builders are focused on vertical AI applications, making application-layer products the most commonly reported AI product among AI offerings.
49% of companies report their primary differentiation comes from application-layer innovation (UX, workflows, integrations, data application), compared to a small minority (14%) relying primarily on proprietary model development.
As base models evolve and become more interchangeable, we believe a competitive advantage is increasingly accruing to companies that deeply understand customer workflows and embed AI directly into mission-critical processes. Model ownership appears to matter far less than product experience and distribution.
.png)
2. Multi-Model Architectures Are Becoming the Standard, Driven by Cost and Control
AI builders are increasingly converging on multi-model strategies to manage tradeoffs between performance, cost, latency, and customization.
Model reliability and accuracy remain the top selection criteria, but cost ranks second, reinforcing what we view as the shift toward cost-efficient stacks.
Builders now use ~3.1 model providers on average, up from ~2.8 six months prior, signaling increasing architectural diversification.
Perspectives shared by the ICONIQ Community support that many companies are routing the majority of workloads to smaller or fine-tuned models, escalating only high-complexity tasks to frontier models. This orchestration approach is increasingly tied to margin outcomes, as the companies surveyed expect gross margins to continue improving, reaching ~52% on average in 2026.
3. Signals of AI Monetization Evolving, But No Final Answer
AI pricing remains an unsettled dimension across the market.
58% of companies still include a subscription or platform component, but consumption-based (35%) and outcome-based (18%) pricing have grown meaningfully over the last six months.
37% of companies plan to change their AI pricing model in the next 12 months, driven primarily by customer demand, competitive pressure, and margin concerns.
Notably, companies experimenting with outcome-based pricing most often tie outcomes to cost savings (36%) or revenue generated (18%), underscoring that monetization is increasingly linked to demonstrable business value rather than feature access alone.
Across the data and interviews, hybrid models (light platform fees plus usage, with safeguards like annual commitments and tiered overages) are emerging as the recommended pragmatic approach amid stabilizing AI economics and customer value curves.
4. AI Is Acting as a Force Multiplier Across Organizations
Internal AI adoption is moving beyond experimentation, with respondents reporting measurable productivity gains across functions.
R&D teams tend to lead adoption rates, with 60% of employees actively adopting AI tools within the function. Use cases such as coding assistance, testing, documentation, and content generation show the highest reported productivity improvements, often exceeding 30 - 40% time savings. However, integration with existing workflows and accuracy of AI models remain top challenges when adopting AI for internal use cases, highlighting the importance of both model selection and change management to accelerate adoption.
As adoption matures, companies are spending a higher percentage of revenue on internal AI tools and measuring ROI through productivity gains, cost savings, and revenue uplift.
Importantly, AI has not yet driven significant reductions in headcount; instead, it seems to be reshaping workforce composition. Companies are prioritizing AI-fluent talent while de-emphasizing administrative and repetitive roles. The data suggests that internal AI is becoming a force multiplier for existing teams, rather than a near-term lever for workforce reduction.
5. The AI Tooling Ecosystem Is Maturing
We surveyed hundreds of companies to understand which tools companies are using for AI product development.
Below, you’ll find a quick tour of the tools most widely reported, listed in alphabetical order.3 If you’re building something different or doubling down on a rising alternative, we’d love to hear from you: ICONIQGrowthInsights@iconiqcapital.com.
1. Perspectives from the ICONIQ GenAI Surveys (April 2025 & December 2025) and perspectives from the ICONIQ team and network of AI leaders consisting of our community of CIO/CDOs overseeing AI initiatives in enterprises, CTOs, our Technical Advisory Board, and others in our network
2. Survey responses include some but not all ICONIQ Venture and Growth portfolio companies as well as companies not part of ICONIQ Venture and Growth’s portfolio. Please refer to the full report for additional information here.
3. Tools shown here represent most commonly used developer tools as selected by survey respondents as of Dec 2025, limited to those chosen by above 2% of total respondents.