The AI Investing Landscape: Insights from Venture Capital

In 2024, the San Francisco Fed and the Federal Reserve System Innovation Office launched the EmergingTech Economic Research Network (EERN) to support a better understanding of how new technologies like GenAI are shaping the economies of today and the future. As part of the EERN initiative, we often hold roundtable discussions to hear from industries and sectors at the front lines of digital adoption to ask questions such as: How are artificial intelligence (AI) tools being used? Is AI adoption still in early stages or has it been fully integrated in certain areas? What are the expected or realized impacts on firm productivity and employment?

To address these and other questions about AI investment trends, we recently convened a roundtable of venture capital investors specializing in AI. These leaders from across the venture capital ecosystem gathered for a discussion on current investment patterns, economic impacts, and talent and resource constraints in the AI space. The venture capitalists met with Mary C. Daly, President and CEO of the San Francisco Fed, Sunayna Tuteja, Senior Vice President and System Chief Innovation Officer for the Federal Reserve System, and Kevin Ortiz and Huyiu Li, co-heads of EERN.

The Maturing AI Investment Landscape

The AI investment ecosystem has evolved significantly over the past 18 months according to roundtable participants, who said that what began as an initial wave of frenzied speculation has given way to a more sophisticated approach where investors increasingly focus on sustainable business models and concrete value creation.

Participants noted that AI investing has matured considerably, with ample capital now available but flowing more strategically, particularly concentrated in major technology hubs. This evolution reflects growing investor sophistication, with an emphasis on real customer usage and credible unit economics rather than speculative potential.

Several investors highlighted their focus on companies developing technological “moats” – sustainable competitive advantages that could include proprietary data, domain-specific technology, superior speed, or established distribution channels. They noted that these defensible positions are increasingly seen as essential for long-term success in the AI space.

From Infrastructure to Applications

The discussion revealed an evolving shift in investment focus within the AI ecosystem. While infrastructure investments remain significant especially by large technology companies, participants shared that many venture capitalists are increasingly turning their attention to application-layer opportunities.

As models have improved and markets gained clearer views of possibilities unlocked by AI, additional capital is flowing into user-facing AI applications. Participants noted growing interest in vertical-specific AI applications targeting regulated and complex industries such as legal, healthcare, financial services, and accounting. These specialized solutions are seen as critical to driving broader adoption by lowering barriers for customers in these sectors.

The evolution of AI infrastructure was also a significant topic of discussion. Participants observed that data centers previously built for cryptocurrency mining are being repurposed for AI and high-performance computing, reflecting substantial demand for computational resources. However, some investors expressed caution about long-term infrastructure plays, noting potential risks if computing eventually shifts more toward edge devices like smart phones.

The AI Talent Equation

Competition for AI talent emerged as a critical challenge facing the ecosystem during the discussion. According to participants, top AI startups face extremely high labor costs, with some specialists commanding substantial compensation packages. This talent premium is creating a bifurcated market where well-funded companies have significant advantages in attracting key personnel.

Roundtable participants observed that AI is enabling new operational structures, with leading startups employing AI agents alongside human talent at ratios as high as 10:1 AI agent to full-time employee. They shared that this approach allows smaller teams to achieve greater output, though participants emphasized that human judgment remains essential, particularly for complex decision-making.

Several venture capitalists highlighted that AI is enabling solo founders and small teams to accomplish what previously required much larger organizations, saying that the technology has fundamentally changed team structures and operational efficiency across many sectors.

Economic Impact and Labor Market Transformation

The potential economic impact of AI sparked thoughtful discussion among the roundtable participants. While bullish on AI’s transformative potential, investors generally expressed nuanced views about its effect on employment and productivity.

Participants emphasized that in their experience AI primarily augments rather than replaces human workers. They described how the technology requires reskilling workers as job functions evolve, with AI serving as a productivity accelerator that enables employees to work more efficiently and effectively rather than replacing them outright.

However, participants also acknowledged concerns about societal adjustment to rapid technological change. They noted the psychological dislocation that can accompany technological transitions, highlighting anxieties many professionals feel about adapting to AI-enhanced workflows and preparing their children for future workforce demands.

The Future of AI Investing

Looking ahead, roundtable participants made clear that there remains significant areas for future AI-focused investing. One participant highlighted the need for further investments in non-language based foundational layers. For example, this participant noted investments in mathematical reasoning capabilities in AI systems could help with theoretical mathematics problems and address current AI-limitations like hallucinations. This investor noted recent academic research showing progress in this area, while acknowledging significant work remains ahead.

The venture capitalists generally agreed that we’re still in early stages of AI adoption, with significant innovations yet to come. This perspective is informing their investment strategies, with some firms delaying certain application funding to instead focus on early-stage research and development expected to bear fruit in 2026 and beyond.

When discussing how they themselves use AI in investment decisions, participants emphasized that while AI tools enable more informed decisions, they don’t replace human judgment. Relationship building and qualitative assessment of companies and their leadership remain fundamentally human activities, with technology serving as an enhancement rather than a replacement.