How are VCs adapting their investment strategies to AI disruption?

Venture capitalists are fundamentally re-evaluating their investment theses to adapt to the AI disruption, moving away from traditional SaaS heuristics toward new frameworks centered on defensibility and capital efficiency [1, 6]. Investors are now prioritizing companies whose competitive moats strengthen as underlying foundation models improve and become commoditized, a key diligence question for firms like Benchmark [1, 15, 28]. The investment focus is shifting from selling software tools to backing companies that replace labor or use AI to "own the outcome" in specific verticals with proprietary data and complex workflows [4, 13]. Some investors have adopted AI as a definitive filter, refusing to invest in companies that are not clear beneficiaries of the technological shift [12, 21, 29]. This has led to a significant reallocation of capital, with traditional software companies struggling to raise funds while AI-native startups receive heightened attention [16, 22]. The value is increasingly seen at the application layer, where startups create composite systems from multiple models, rather than in the foundational models themselves [10, 24].

A central tension exists among investors regarding whether incumbents or startups will be the primary beneficiaries of the AI wave. One perspective holds that established companies are best positioned to win by leveraging AI to enhance productivity and capitalize on existing distribution networks and data advantages [2, 11]. Proponents of this view are selectively buying publicly traded software stocks at low valuations, betting on their ability to adapt and arguing that defensible "systems of record" like ERPs and CRMs will endure [25, 27]. Conversely, other prominent VCs view AI as a new computing platform on par with the internet, predicting it will generate an unprecedented number of new billion-dollar companies . This camp believes the biggest winners will be today's startups, drawing parallels to the emergence of Google and Amazon after the dot-com bubble, and that incumbents will be forced into a wave of M&A to acquire AI capabilities and remain competitive [10, 30].

The current investment landscape is characterized by a speculative bubble dynamic, forcing VCs to navigate high valuations and novel deal structures [4, 7]. The environment is challenging traditional, disciplined approaches, with high-valuation, pre-revenue bets becoming more common . Concerns are rising about financial transparency due to "circular deals," where corporate VCs invest in startups that then use the capital to purchase cloud compute or chips from the investor's parent company, potentially inflating valuations . In response, VCs are showing flexibility on entry valuations for top-tier companies but remaining firm on ownership targets to align with their support models . This market shift is also changing financial evaluation metrics; the classic "triple, triple, double, double" SaaS growth model is now considered obsolete, replaced by the potential for explosive, non-linear adoption curves seen in successful AI products [5, 9]. Investors are also recalibrating their financial models to account for AI-native companies having structurally lower gross margins due to inference costs, with the expectation that this will be offset by higher terminal operating margins from greater operational efficiency .