Early seed investing, now a core part of the venture ecosystem, began with small, experimental funds like Jeff Clavier's $15M fund. This was considered perplexingly small at a time when larger VC funds dominated, highlighting the contrarian origins of today's seed stage.
Even highly successful accelerator cohorts, like a Techstars class from 2014-15 that produced two unicorns, are still waiting for exits nearly a decade later. This underscores the reality of prolonged holding periods in venture capital, where even top-tier outcomes can take over 10 years to materialize.
Frontier AI model providers like OpenAI and Anthropic are in discussions for a regulatory deal. They would receive a product liability shield, similar to Section 230 for platforms, in exchange for contributing 5-20% of their equity to a new government-managed sovereign wealth fund.
The most immediate threat from AI is not a rogue superintelligence but human over-reliance on the technology. As people increasingly trust AI outputs without critical thought, their cognitive abilities may decline, making them more likely to rubber-stamp potentially catastrophic AI-generated errors in critical applications.
Rather than making people less capable, AI serves as a powerful equalizer. It provides "total unlocks" for individuals in areas where they lack innate talent, such as a technical founder using AI to build business models or a non-writer drafting compelling copy. This democratizes access to professional capabilities.
The strategic rationale behind Stripe's acquisition of AI router OpenRouter is likely not about optimizing LLM costs. Instead, Stripe is leveraging its core competency in anti-fraud to build a trust and safety layer for AI, positioning itself to monitor and prevent malicious use of AI models.
Startups like AI-assistant Instinct create massive hype and high-velocity funding rounds through a "VC baiting" playbook, reminiscent of Clubhouse. By giving early access exclusively to the venture community, they generate intense FOMO and competitive tension among a few top funds, driving up valuations before achieving public traction.
A significant warning sign for consumer AI startups is when they burn excessive cash on compute while keeping the product on a limited waitlist. This suggests either a fundamental engineering inefficiency, an unsustainably complex tech stack, or a flawed unit-economic model that prevents them from scaling.
Faced with a price war and the growing dominance of open-weight models, venture capitalists are shifting their investment thesis. They see more durable value in the open-source ecosystem and enabling infrastructure than in closed frontier models, which they view as a commoditizing "race to the bottom."
In today's high-valuation environment, elite seed funds like Uncork Capital have adapted. Instead of avoiding high prices, they pay the market rate for exceptional companies but write larger checks to secure their target 10-12% ownership. This discipline requires every investment to have multi-billion-dollar potential.
The goalposts for going public have moved dramatically. A CEO of a private company already doing over $500M in revenue stated they wouldn't consider an IPO until reaching $10 billion in revenue, not valuation. This reflects a desire to avoid public market scrutiny until achieving massive scale.
The fundraising market has shifted dramatically. A startup at the intersection of fintech and healthcare with $3M ARR and strong fundamentals was unable to raise a Series A. Investors passed due to the categories not being "sexy," forcing the founders to raise a smaller seed extension despite their traction.
In the past, achieving the "Triple, Triple, Double, Double, Double" (T2D3) revenue growth rate was a near-guarantee for a successful fundraise. Today, that is no longer the case. Investors have become desensitized to strong fundamental growth, prioritizing hype cycles and making funding unpredictable even for top performers.
Venture funds can create liquidity for their Limited Partners (LPs) by selling a portion of a top-performing company from an older fund into a new, self-managed Special Purpose Vehicle (SPV). This generates DPI for LPs who want to cash out, while allowing others to roll into the new SPV and retain exposure.