A vision for AI and the future: How Boldstart Ventures seeks to fund the next wave of enterprise innovation

AI

The stage is set for a wealth creation cycle that we have never seen before.

The roles played by data and AI highlight the instant use cases of a generational shift in how things are produced and consumed. Power dynamics are shifting to the founders of exciting new startups that are well-positioned to take advantage of this wave of wealth creation and innovation.

To gain a better understanding of how this will evolve, I sat down recently for an exclusive interview with Ed Sim (pictured), founder and general partner at Boldstart Ventures Management LLC, at theCUBE’s Palo Alto studio.

As a prolific enterprise investor and newsletter writer (whatshotit.vc), Sim has witnessed the cycles of the tech industry for more than two decades. He opened our conversation by noting that the current wealth creation period could be historic.

“I think this will be the best class of startups that comes out,” Sim said. “You’re going to look back at this and be like, ‘Oh my god, there were some amazing founders that built some amazing companies that were really efficient in leveraging this AI thing.’”

Funding wave for AI

The influence of AI is already having a significant impact within the investment community. Generative AI startup Anthropic PBC is reportedly in talks to raise $750 million on a $15 billion valuation, and the large language model startup Cohere Inc. is holding discussions with investors to raise between $500 million and $1 billion in fresh funding.

This is being driven by enterprise demand for AI tools that can be rapidly assimilated into the business, and it is powering many of the enterprise models that Sim sees today.

“If you’re not thinking about how to leverage AI in your product, whether it’s just a feature or maybe a new product offering, then I think you’re insane,” Sim said. “The technology with the LLMs out there, the intelligence around it, the ability to use an API or even grab an open-source model and train it to give you an instant result, I think it’s off the chart. Cost compression has come down a lot, and most founders I see today are starting with some type of AI leveraged product.”

The question now confronting many enterprises today is whether to choose a proprietary AI solution or embrace one of a growing number of open-source models. The management issues that surrounded OpenAI in November will likely result in a mixed-choice scenario, according to Sim.

“After you saw the governance issues that arose last year … I think people understand that to have one company control access and APIs to all enterprise applications is a scary place,” Sim explained. “That really opened up the opportunity for the rise of open source. We’re going to live in a multi-model world, where there is going to be some open source and some OpenAI, depending on what cost issues you may have and what privacy issues you may have.”

Pursuing disruptive models

Much as social media companies had to build their platforms from scratch because there was no general-purpose enterprise software to power this new breed of applications, the growth of AI presents a similar opportunity. Sim hinted that Boldstart Ventures is pursuing at least one such disruptive model in the supply chain management arena.

“No matter how big a company is, it’s open for reinvention,” Sim said. “We backed some founders, and I won’t say who, but they were looking at the SCM market. When was the last time we entered innovation around GitHub or GitLab? Of course, there’s some AI-related angle to that, let’s just say. When you take a new technology as groundbreaking as AI, you can actually build it from scratch in a better way than bolting on AI on top of an existing infrastructure.”

As with any evolving technology, there are still challenges involved in AI adoption. One of these involves consistency in the technology’s application across a broad range of enterprise use cases.

“The hard part right now when you’re looking at AI in the enterprise is that last-mile piece,” Sim said. “It’s easy to get up and running, to call an API and get an answer. The question is: Can you get that answer repeatable?”

Another question surrounds the ability of AI tools to transverse environments at the infrastructure level and run in distributed fashion. Sim and his venture firm are paying close attention to this and its implication for growth at the edge.

“Think about the privacy and data implications … the cost implications, being able to process and give you answers on the edge on these devices without reporting back to the cloud,” Sim said. “Think about all of the opportunities like sensors sitting out, collecting data on windmills. Automobiles are moving data centers right now. There’s just going to be an explosion of opportunities looking at edge related stuff over time.”

Sim and his partners at Boldstart have followed a philosophy of working with founders based on a strategy of “inception investing.” It’s an approach that engages with founders well before they incorporate to battle test and iterate on their ideas before going to market. That takes trust and a vision for the future.

“We love infrastructure, we love dev tools, we love cyber, but we don’t see the future,” Sim said. “I need to trust you to see the future. We want to help founders accelerate their path to product market fit and avoid the mistakes that we’ve seen for 28 years.”

Here’s theCUBE’s complete video interview with Ed Sim, one of many more exclusive CUBE Conversations:

Photo: SiliconANGLE