Shanghai | WAIC 2026
For much of the past three years, artificial intelligence has been valued on promise.
Investors rewarded companies for larger foundation models, faster chips, higher token consumption and ambitious visions of Artificial General Intelligence (AGI). Revenue multiples expanded rapidly while profitability often remained a secondary consideration.
WAIC 2026 suggests that this valuation framework is beginning to evolve.
Walking through the exhibition halls in Shanghai, one observation became increasingly clear: AI is no longer being presented primarily as a technological breakthrough. It is being presented as productive infrastructure.
Humanoid robots are assembling electronic components, managing warehouse logistics and supporting retail operations. Enterprise AI agents are automating customer engagement, knowledge management and operational workflows.
The industry’s conversation has shifted from what AI can generate to what AI can deliver.
That distinction matters for investors.
Deployment, Not Demonstration
One of the most notable differences at this year’s WAIC was the disappearance of technology demonstrations designed purely to impress audiences.
Instead of robots dancing or performing choreographed movements, exhibitors showcased deployment.
Factories.
Hospitals.
Warehouses.
Retail stores.
Hotels.
Distribution centers.
Nearly every demonstration revolved around operational efficiency rather than technical novelty.
This evolution mirrors what has happened across previous technology cycles.
Cloud computing became valuable not because virtual machines existed, but because enterprises rebuilt their operations around them.
Similarly, generative AI may become economically significant not because models continue to improve, but because enterprises successfully integrate AI into everyday business processes.
Commercial adoption—not model performance—is increasingly becoming the industry’s defining metric.
Enterprise Knowledge Is Becoming the New Competitive Advantage
Large language models have significantly lowered the cost of intelligence.
What they have not replaced is enterprise knowledge.
Every large organization possesses unique operational expertise accumulated over years—or decades—through customer interactions, supply chains, compliance frameworks and industry experience.
These proprietary knowledge assets cannot simply be downloaded from the public internet.
Increasingly, competitive advantage lies in connecting foundation models with enterprise-specific knowledge and workflows.
This explains why enterprise AI platforms are attracting growing attention.
Rather than competing directly with foundation model providers, enterprise AI companies focus on enabling organizations to transform internal knowledge into deployable AI capabilities.
In many cases, the value does not come from creating another model.
It comes from making existing models useful inside complex businesses.
AI Infrastructure Is Expanding Beyond Compute
Earlier stages of AI investment centered largely on semiconductors and cloud infrastructure.
That infrastructure remains essential.
Yet WAIC 2026 highlighted another layer gaining strategic importance.
Enterprise AI infrastructure.
This includes:
enterprise knowledge platforms;
intelligent agent orchestration;
workflow automation;
security and governance;
industry-specific AI deployment.
These capabilities increasingly determine whether AI creates measurable business outcomes.
Infrastructure is no longer defined only by GPUs and data centers.
It is also defined by the software architecture enabling AI to operate reliably inside enterprises.
Marketingforce Reflects This Industry Transition
Marketingforce (HKEX: 2556.HK) represents one example of this broader shift toward enterprise AI deployment.
Rather than positioning itself as a foundation model developer, the company focuses on integrating AI infrastructure, enterprise knowledge management and intelligent agent applications into a unified platform designed for business customers.
Its strategic cooperation with China Telecom to jointly develop AI computing infrastructure and an AI application innovation center further reflects an ecosystem approach to enterprise AI deployment.
The combination of computing resources, AI platforms, enterprise knowledge and intelligent agents illustrates how commercialization increasingly depends on integrated infrastructure rather than isolated technologies.
Why Profitability Matters Again
Perhaps the most important implication of WAIC 2026 is financial rather than technological.
As enterprise AI adoption accelerates, investors are beginning to ask different questions.
Instead of asking:
“How large could this market become?”
They increasingly ask:
“Can this business generate sustainable earnings?”
That represents a significant shift in valuation methodology.
Earlier AI leaders were frequently valued primarily on revenue growth.
Today, operating leverage, recurring enterprise adoption, cash generation and profitability are becoming equally important.
Marketingforce’s recent earnings guidance, projecting attributable net profit growth of approximately 386% to 494% for the first half of 2026, reflects this broader transition toward commercially scalable AI businesses.
While individual company performance should always be assessed within its broader operating context, the result illustrates an important industry trend: enterprise AI companies are beginning to demonstrate operating leverage as deployment scales.
From Price-to-Sales Toward Price-to-Earnings
The AI investment narrative is unlikely to abandon growth.
Growth remains essential.
However, valuation frameworks are becoming more balanced.
Revenue multiples may continue to matter for early-stage AI innovators.
Yet companies capable of converting AI adoption into sustainable earnings could increasingly be evaluated through profitability metrics alongside growth.
The conversation is gradually moving:
From model capability…
to commercial capability.
From token consumption…
to customer value creation.
From infrastructure investment…
to enterprise productivity.
And ultimately,
from Price-to-Sales toward a combination of Price-to-Sales and Price-to-Earnings.
The Next Phase of AI
WAIC 2026 did not signal the end of the AI innovation cycle.
It signaled the beginning of a more demanding one.
The companies most likely to succeed over the next decade may not simply build more capable models.
They will build businesses that help enterprises deploy those models safely, efficiently and profitably.
As AI becomes embedded within operational workflows rather than remaining a standalone technology, value creation will increasingly depend on execution.
The next chapter of artificial intelligence may therefore be less about bigger models—and more about better businesses.