Artificial intelligence is providing a significant boost to Silicon Valley's earliest-stage companies. Startup accelerator Y Combinator (YC), known for backing successful ventures like Airbnb, Dropbox, and Stripe, recently hosted its annual demo day in San Francisco. During the event, startup founders pitched their companies to potential venture capital investors.

Garry Tan, CEO of Y Combinator, highlighted the impressive growth rate of the current cohort of startups. He noted that the group is growing significantly faster and generating actual revenue at a higher pace than previous cohorts. Over the past nine months, the entire batch of YC companies has collectively grown by 10% per week. Tan emphasized that this growth extends beyond just a few top companies, indicating a broader trend across the entire batch. He attributed this growth spurt to advancements in artificial intelligence.

According to Tan, app developers can now leverage AI to automate repetitive tasks and generate new code using large language models. He introduced the term "vibe coding" to describe the process of allowing AI models to take the lead in software generation. In some cases, AI can even code entire applications. By subsidizing traditionally heavy workloads, AI enables these companies to operate with smaller teams. Tan reported that approximately 25% of the current YC startups have had AI write 95% of their code.

While the idea of AI writing most of the code may seem concerning, it offers founders the advantage of not requiring large engineering teams. Tan stated that companies are achieving revenues of up to $10 million with teams of fewer than ten people. This efficiency reduces the need for extensive capital raising and extends the lifespan of available funds. The growth-at-all-costs mentality that prevailed in Silicon Valley during the zero-interest-rate era has diminished, with a renewed emphasis on profitability.

Tan also pointed out that this focus on the bottom line extends to megacap tech companies. Companies like Google, Meta, and Amazon have undergone multiple rounds of layoffs and reduced hiring. While this has caused anxiety among some engineers, Tan views it as an opportunity for startups. It has become easier to build a startup, and top talent no longer feels compelled to work for big tech companies to prove their worth.

Tan believes that engineers who may not have secured positions at Meta or Google now have the opportunity to build standalone businesses generating substantial revenue with smaller teams. Approximately 80% of the YC companies that presented this week are focused on AI, with a smaller number working on robotics and semiconductors. Tan noted that these companies have demonstrated commercial viability earlier than previous generations.

He acknowledged the hype surrounding AI but emphasized the unique aspect of this moment: real commercial validation. Investors at demo day can contact actual customers who confirm their daily use of the startups' software.

Y Combinator, founded in 2005 by Paul Graham, Jessica Livingston, Robert Morris, and Trevor Blackwell, invests $500,000 in startups in exchange for an equity stake. Founders then participate in a three-month program at the San Francisco headquarters, receiving guidance from partners and YC alumni. Demo day serves as a platform to attract additional capital.

The firm has funded over 5,300 companies, with a total valuation exceeding $800 billion. More than a dozen of these companies are publicly traded, and over 100 are valued at $1 billion or more. With an acceptance rate of approximately 1%, Y Combinator receives over 15,000 applications from companies seeking to join the accelerator.

While the number of venture capital incubators has increased in recent years, along with capital flowing to early-stage startups, Tan argues that Y Combinator maintains an edge due to its strong network. He highlighted the increasing number of highly valued portfolio companies and refuted the notion that specialized incubators are taking business away.

Tan explained that around 20% to 30% of companies within YC change their initial idea, sometimes even their entire industry. He cautioned that specialized incubators may limit the ability to pivot into the most promising direction. He concluded that the network effects and advantages of participating in YC have only become more pronounced.