AI seed startups face higher valuations and faster growth demands

The market for AI seed startups has shifted, with investors pricing rounds years ahead of traction. According to founders and VCs, seed deal count is down but valuations are up, as investors show little interest in companies outside of AI. Ashley Smith, a general partner at the early-stage fund Vermilion, stated that investors in this market are pricing rounds years ahead of traction.

To illustrate this shift, consider the valuation changes between 2024 and today:

• In 2024: Pete Martin, founder of Realm, raised a $5 million seed round at a $25 million post-money valuation (the company’s valuation after an investment is made).
• Today: A typical seed round is $10 million at a post-money valuation of $40 million to $45 million.

This pressure was visible at the Y Combinator Demo Day in March, where some startups asked for $5 million at a $40 million post-money valuation. Smith noted that this pricing went beyond the YC tax—the premium paid for startups that went through the Y Combinator accelerator.

The valuation surge is driven by pressure to demonstrate fast traction. Shanea Leven, founder of Empromptu, pointed to Cursor—which hit $100 million in revenue in 12 months in early 2025—as a driver of these expectations. Leven noted that investors now expect this performance, adding that the pressure is at an all-time high to build not just a billion-dollar company, but a $50 billion one.

To secure access to deals, venture firms are adjusting their strategies. Some are moving into pre-seed rounds—the very early stage of startup funding—while others are increasing check sizes. Marlon Nichols, managing general partner at MaC Ventures, said his firm’s average entry check has grown from $1 million in 2019 to $2.5 million today, capping at around $5 million. Nichols’ last two seed investments, which valued the startups at $25 million and $30 million post-money, respectively, were already generating $2 million in revenue and involved checks between $3 million and $4 million.

Meanwhile, Amber Atherton, a partner at the early-stage consumer fund Patron, reported that the average check size for the firm’s $100 million Fund II ranges from $4 million to $5 million, up from $1 to $2 million for Fund I. Jonathan Lehr, a general partner at Work-Bench, which invests out of a $160 million fund, noted that his firm has become increasingly comfortable investing at the pre-seed stage as companies scale faster.

This environment is also shaped by competition for AI talent. “There’s a war for great researchers right now, and I don’t think it’s good or bad; it’s just the current state of the market,” said Amber Atherton, a partner at Patron. This talent competition has driven valuations like the $2 billion seed round for Thinking Machine Labs at a $12 billion valuation.

However, this environment leaves less room for error. Lehr warned that higher seed valuations mean less margin for error, less room for experimentation, and less tolerance for pivots.

Why it matters

The surge in AI seed valuations is fundamentally changing the venture capital playbook, forcing investors to bet on early traction and talent rather than just ideas, while leaving founders with less margin for error.