Which markets do VCs feel are the most interesting to invest in today, and which are the hardest to invest in?

Venture capital investors are executing a significant capital rotation away from the software sector that dominated the last decade and toward physical assets and infrastructure [3, 9, 12, 18]. This shift is driven by the demands of the AI boom, with investors like Summit Partners focusing on the "physical layer" of technology, including photonics and AI-enabling semiconductors . Other areas of interest include energy and data centers [3, 12]. This reallocation has created a difficult fundraising environment for traditional software companies, which are struggling to raise capital amidst the AI craze . The sector's fall from favor is so pronounced that in the leveraged finance market, software has reportedly gone from a leading category to being **almost unfinanceable** [7, 8]. This broad consensus points to a fundamental change in the venture landscape, where the tangible underpinnings of technology are now valued over the pure-play software models of the previous cycle.

The AI sector itself presents a complex and bifurcated investment landscape, characterized by both intense interest and significant skepticism. On one hand, institutional investors are reported to have "zero interest" in startups not related to AI , with leading VCs adapting their theses to target companies that replace human labor rather than simply sell software [11, 25]. The most interesting technical frontier is seen as models based on pure reinforcement learning . On the other hand, some investors believe the odds of generating a **100x or greater return** on a new AI investment are now extremely low, as the most successful entries were made before the current hype cycle . The frontier model market is particularly challenging due to heavy subsidization from large corporations . A more patient, cyclical view suggests the current AI wave is a bubble, and the most valuable "Gen 2" companies will be founded in the next 2-5 years, following a significant market downturn that rationalizes valuations [1, 4].

Beyond the dominant AI narrative, investors are identifying contrarian opportunities in overlooked domestic sectors and international markets. The intense focus on AI and SaaS has created a vacuum in consumer internet, presenting a significant opportunity for investors willing to fund businesses in a less competitive landscape [16, 23]. Geographically, China is described as a **"capital starved"** and "least crowded trade," with specific opportunities in robotics . This view of China as an underfunded market exists in tension with the prediction that it will ultimately win the global AI race due to its ingenuity and fewer infrastructure hurdles . Other international markets of interest include India, where the venture exit market is expected to mature over the next decade , and Europe, where the war in Ukraine has made defense technology a newly necessary and acceptable area for venture investment .

Several other sectors are experiencing difficult funding environments, mirroring the challenges in traditional software. In biotech, venture investors are concentrating capital on a smaller number of existing portfolio companies, making it exceptionally difficult for new or struggling firms to raise funds [19, 22, 26]. This has led to a large number of "zombie" biotechs trading for less than their cash on hand . The data streaming market is also cited as a historically difficult category for venture investment, with very few breakout successes . These sector-specific struggles are compounded by a challenging macro environment for LPs, who are navigating a liquidity crunch and the widely expected underperformance of venture fund vintages from 2020 and 2021, which were deployed at peak valuations [2, 5, 30].