AI investment shifts toward "real demand": Investors' screening logic for startups is changing

AI Investment Shifts Toward 'Real Demand': Investors' Screening Logic for Startups Is Changing

A conversation about the future of AI entrepreneurship is unfolding in Silicon Valley. At the closely watched TechCrunch Disrupt conference, investors sent a clear signal: AI remains capital's most certain bet, but the investment logic is undergoing a subtle and profound restructuring.

Over the past year, the explosion of generative AI has led nearly every startup to include the word "AI" in its narrative. However, as the hype fades, investors are scrutinizing projects with a more discerning eye. Several investors at the conference—including Nina Achadjian, Jerry Chen, and Peter Deng—shared a common assessment in a discussion about future investment directions: AI is no longer a field where one can secure funding based on grand concepts. Startups must prove they have the resilience to survive in a rapidly changing market and deep domain expertise in specific industries.

From 'Storytelling' to 'Problem Solving'

In the early stages of AI investing, an impressive demo or a cool model was enough to attract funding. But today, enterprise customers have grown weary of vague AI promises. Investors are increasingly focusing on whether startups genuinely solve real problems, rather than merely offering a technology demonstration.

This means founding teams need a deep understanding of their target industries. The complex processes in verticals such as healthcare, finance, and manufacturing cannot be easily penetrated by a general-purpose AI model. Companies that can deeply integrate AI technology with industry knowledge are becoming the targets of capital.

Data Strategy Becomes the New Moat

Another notable change is that data strategy has replaced model parameters as one of the core indicators investors use to evaluate AI startups. Against the backdrop of increasingly homogenized foundational model capabilities, unique, high-quality data sources and compliant data processing capabilities form the real competitive barriers.

Investors unanimously mentioned that founders need to clearly present their data strategy: where to obtain data, how to ensure data quality, and how to build a data flywheel. This is not just a technical issue, but a matter of business model and sustainable competition.

Real Application Scenarios Emerge

After more than a year of market validation, AI's killer applications are beginning to concentrate in a few specific areas. Chat apps remain an important traffic gateway, coding tools have been widely adopted by developers and enterprises, and intelligent customer service platforms have become an intuitive choice for enterprises to reduce costs and improve efficiency.These applications share a common trait: they solve clear, high-frequency, and quantifiable business pain points. In contrast, projects that try to reinvent everything with AI but lack specific scenarios are losing investors' patience.

Investors also pointed out that AI has long-term potential in areas such as e-commerce marketplaces and robotics. In particular, the combination of AI with the physical world—from warehouse automation to on-site work in blue-collar industries—is seen as the next wave of structural opportunity.

Digitization of Blue-Collar Industries: An Underestimated Opportunity

The discussion specifically noted that digitizing traditional manual processes is an underdeveloped goldmine in investors' eyes. In blue-collar industries such as construction, logistics, and manufacturing, a large amount of work still relies on paper forms and manual coordination, and AI-driven digital solutions have the opportunity to significantly improve efficiency.

This field is underestimated because it is often less eye-catching than consumer-grade AI products, but it is massive in scale and customers have a clear willingness to pay. For entrepreneurs who can go deep into these industries and understand the pain points of frontline workers, this may be a broader track.

The New Normal of AI Investment from a Global Perspective

From a global perspective, AI investment is moving from "casting a wide net" into a phase of "intensive cultivation." Not only Silicon Valley, but capital from Europe, the Middle East, and Asia is also pouring into the AI track at an unprecedented pace, but fund screening criteria are tightening in tandem. Entrepreneurs need to realize that under the dual pressures of intensifying market competition and increasingly rational capital, relying solely on the "AI concept" is no longer viable.

For startups, the core question they need to answer now is no longer "what can you do with AI," but "why can you survive the AI wave and build a lasting advantage." The answer is likely hidden in data, industry expertise, and keen insight into real needs.

This article is written based on public discussions at the TechCrunch Disrupt conference, aiming to provide industry trend analysis and does not constitute investment advice.