Insight Partners diversifies AI bets against OpenAI and Anthropic concentration

Insight Partners is taking a contrarian stance in the current venture capital landscape, explicitly rejecting the strategy of concentrating funds into a small number of frontier AI labs. According to an interview with TechCrunch at its StrictlyVC event in New York, co-founder Deven Parekh stated that while OpenAI and Anthropic have absorbed a massive share of recent venture dollars, Insight Partners has maintained a diversified portfolio to mitigate concentration risk. Parekh noted that while the firm holds stakes in both OpenAI and Anthropic, it does not view them as the sole drivers of its returns, contrasting this with peers who are raising funds with pitches centered on allocating 35-40% of capital to a single company.

The case for early-stage entry

Parekh argues that the current market environment, characterized by rapidly rising valuations and fast-moving rounds, favors investors who enter earlier in a company’s lifecycle. He observed that in the current cycle, follow-on rounds often occur without significant incremental data, meaning investors are paying higher prices without a corresponding reduction in risk. "The logical response is to go earlier," Parekh said. He explained that by writing smaller checks, such as $20-25 million, rather than massive $500 million late-stage rounds, Insight Partners can make more bets and double down on winners. He cited the firm’s investment in Wiz as a prime example, noting that by writing a Series A check and continuing to invest, Insight’s gains were significantly larger than if it had stopped at the first round. This strategy allows the firm to manage downside risk more effectively, as a failed early-stage bet has a minimal impact on a fund of Insight’s size.

Navigating conflicts of interest

Investing in direct competitors like OpenAI and Anthropic was once considered taboo in venture capital, but Parekh says the dynamics have shifted. He explained that the internal debate at Insight Partners was less about the conflict of interest and more about whether the firm had missed the opportunity to invest in earlier rounds. "Once you’re at a later stage, off the board, not driving governance, you’re just buying a great stock," Parekh said. He noted that at the Series A and B stage, the firm adheres to strict information-sharing restrictions and avoids investing in directly competing companies. However, as OpenAI and Anthropic have scaled to raise tens of billions of dollars, they have lost the ability to dictate exclusivity, allowing multiple investors to hold positions. Parekh described OpenAI as the dominant consumer play and Anthropic as having a clear enterprise strategy, though he acknowledged that these positions are shifting in real time.

Global talent and geographic strategy

Insight Partners operates globally, with no fixed geographic allocation. Parekh noted that while AI infrastructure talent is heavily concentrated in San Francisco, vertical AI investing allows for more geographic diversity. He cited the firm’s competition for the AI legal-tech company Legora, where partner Jeff Horing flew to Stockholm to pitch the founder, ultimately losing the deal to General Catalyst. Parekh said the loss was a matter of sales execution rather than strategy, noting, "It’s a big world; we don’t need to win every deal." He also highlighted that talent for vertical AI, such as financial services, is concentrated in other hubs like New York, where companies like Ramp are based. This global approach allows Insight Partners to access a wider pool of founders and mitigate the risks associated with a single geographic market.

Liquidity and DPI

Parekh emphasized the importance of DPI (distributions to paid-in capital) and liquidity, noting that many funds raised between 2021 and 2023 have not returned capital to limited partners. He advised fund managers to take their basis out of positions, even if they believe the company will continue to grow, because LPs need to see cash returns. "LPs want to know you can turn positions into cash; that’s the job," Parekh said. Insight Partners has returned more than $20 billion to LPs through strategic sales and IPOs over the last two years, with a few billion more expected. Parekh noted that secondaries are increasingly attractive as a liquidity mechanism, particularly for early venture investors, and that the firm is actively considering these opportunities to manage its portfolio’s liquidity profile.

What it means for developers

For developers and technical founders, Insight Partners’ strategy suggests a continued appetite for early-stage AI companies that can demonstrate clear product-market fit, even in a crowded market. The firm’s focus on vertical AI and its willingness to invest in companies outside of the San Francisco Bay Area indicate that geographic location is less of a barrier to funding than it was in previous cycles. However, the emphasis on early-stage entry and diversification means that founders may face more scrutiny on their unit economics and scalability, as investors are less willing to pay premium valuations for late-stage rounds without significant incremental data. Developers should also be aware that Insight Partners’ approach to conflicts of interest, while allowing investment in competitors, still involves strict information-sharing restrictions at the early stage, which may impact how founders navigate investor relationships.

What to watch

• Valuation trends: Whether the current pace of rising valuations continues or corrects, which will impact the attractiveness of early-stage entry.
• LP demand for liquidity: How other VC firms respond to the pressure to return capital, potentially leading to more secondaries activity.
• Vertical AI adoption: The pace of adoption in specific verticals, such as legal-tech and financial services, which will determine the success of companies like Legora and Ramp.
• Frontier lab concentration: Whether the high concentration of VC dollars in OpenAI and Anthropic leads to a correction or a continued dominance, which will impact the broader AI investment landscape.