Fortifying Enterprise Security: DynamoFL Secures $15.1M to Combat Data Leaks

0:00What can 160 years of experience teach you about the future? When it comes to protecting what matters, Pacific Life provides life insurance, retirement income, and employee benefits for people and businesses building a more confident tomorrow. Strategies rooted in strength and backed by experience. Ask a financial professional how Pacific Life can help you today. Pacific Life Insurance Company, Omaha, Nebraska, and in New York. Pacific Life and Annuity, Phoenix, Arizona. The wait is over. Dive into Audible's most anticipated collection. The Best of 2025, featuring top audiobooks, podcasts, and originals across all genres.

0:41Our editors have carefully curated this year's must-listens, from brilliant hidden gems to the buzziest new releases. Every title in this collection has earned its spot. This is your go-to for the absolute best in 2025 audio entertainment. Whether you love thrillers, romance or nonfiction, your next favorite listen awaits. Discover why there's more to imagine when you listen at audible.com slash best of the year. Since the beginning of the most recent AI boom, a problem that a lot of companies have faced is security. So right now there is a company called DynamoFL that has just secured$15.1 million in a Series A to fortify enterprise LLMs against data security risks.

1:26Today on the podcast, we'll be breaking down what DynamoFL is going to be doing with that money, why they're raising it, and where we see this going in the future. So the headline here essentially is that Dynamo is a front runner in providing software solutions that essentially integrate large language models with enterprise systems and refine them on sensitive data. And it has made a bunch of waves recently with a$15 million Series A that they just closed. So the really impressive round, I think, was joined by Canopy Ventures, Nexus Venture Partners. They also had some additional contributions from Formas Capital and Soma Capital.

2:01Dynamo FL's overall funding is now at$19 million that they've done to date. And really the vision behind the fund influx, as shared by the company's CEO, is to essentially bolster Dynamo FL's product suite and then amplify its force of privacy research. So in a recent correspondence, he said DynamoFL's product offering allows enterprises to develop private and compliant LLM solutions without compromising on performance. So the genesis of San Francisco-based DynamoFL can really be traced back to 2001 when MIT electrical engineer and computer scientist alumni, Vykunth, and also Christian Lau saw a really pressing need to tackle critical data security flaws that are really just inherent in AI models.

2:50So as generative AI has evolved since then, new concerns have definitely arisen, particularly the fact that LLM's capacity to, you know, remember sensitive training data potentially makes it available to malicious entities, right? So if, you know, LLM is taking in inputs from a client and is remembering them, perhaps using it to train or for other purposes, a malicious actor could ask it, you know, could ask the same LLM about that information and essentially get someone's private information through, you know, that kind of attack. So Mungunthan, who is the CEO, elaborates on the challenge and says, quote, enterprises have found themselves lacking the resources to handle these risks.

3:32Addressing LLM vulnerabilities effectively demands the onboarding of a dedicated team of privacy machine learning experts. What is their role to defies a robust infrastructure to perpetually elevate the LLMs against emerging data security pitfalls, right? So this is definitely a big problem. So I think the corporate sphere is really grappling with LLM adoption obstacles, primarily centered on compliance. Fears are definitely rampant right now about confidential data being accessed by developers who harness you know users data to train models and this apprehension is underscored by decisions from some major giants like apple walmart and verizon who have all to date prohibited their employees from using platforms like chat gpt due to you know these similar kinds of issues so i think what's really interesting is that there was actually just a recent gartner report and i think it shed some light on a few pivotal legal and compliance threats linked with LLMs.

4:32And among them was, you know, of course, the inaccuracy of LLM responses. That's just, you know, general. Everyone knows that these things are not perfect. Concerns over data privacy, confidentiality, model bias. Notably, I think the report underscores the fluctuating regulations based on geographic location, which kind of adds another layer of complexity, right? The EU has rolled out regulations on AI use in, you know, their member countries. And China has recently rolled it out. The United States is, you know, eyeing a bunch of different regulations. So I think that's definitely one that it's kind of hard to grapple with because it just depends based off of, you know, geographically where you are on how that is going to play out.

5:14So DynamoFL's solution is designed for deployment on a client's virtual private cloud or in-house infrastructure. So really they offer a whole bunch of tools to navigate a lot of these, you know, what you might call treacherous waters, right? It's getting kind of crazy out there. And one such tool is the LLM penetration testing instrument. So this was created to identify and log potential LLM data security gaps. And its utility becomes evident when studies indicate that LLMs contingent on their training nuances can unintentionally reveal personal details. So this is a scenario that's a major a red flag for corporations managing exclusive data.

5:55You know, Chase Bank was one of the first companies that was like, hey, no one in our company can use ChatGPT because of issues like this. So this is definitely something that a lot of people in the enterprise are thinking about. Furthermore, DynamoFL's LLM development platform is a beacon of innovation, some people would say, because essentially it is infusing methods to decrease the risk associated with model data leakage and security weak spot. So the platform facilitates developers in integrating a whole host of optimizations, making it feasible for models to operate in hardware restricted setups like mobile phones and edge servers.

6:30So I think it's essential to highlight that while these capabilities sound really groundbreaking, other startups are also doing this, including OctoML, Celdon, and Desi. And OctoML, I specifically have, you know, spoken to some members from their team, they're doing some interesting things over there. So it is important to know that those companies are also working on this. And they also offer tools to refine AI model performance across, you know, different hardware. So some like Lama Index and Contextual AI have directed their energies towards privacy and compliance. So it kind of begs the question, like what makes Dynamo FL stand apart?

7:07So the CEO believes that it's their holistic approach, which is enriched by collaboration with legal experts to ensure DynamoFL's solutions are in sync with privacy laws spanning the US, Europe, and Asia. And I think this really meticulous strategy has gotten the attention of a whole bunch of Fortune 500 corporations, especially from the finance, electronics, insurance, and automotive domains. So the CEO has pointed out the gaps in current market offerings saying, you know, while tools do exist to redact personal information from LLM service queries, they often fall short in sectors like financial services and insurance.

7:46So here, the redacted data can frequently be re-identified through advanced malicious attempts, end quote. So he takes pride in DynamoFL's commitment to building the most exhaustive solution for businesses aiming to fulfill the stringent regulation standards concerning LLM data security. Yet I do think that it is, you know, worth noting that DynamoFL currently doesn't actually cater to the nuanced challenges around IP and also copyright that, you know, LLMs present. So there, you know, this isn't like a one fix for all the issues, but this does, it does offer a pretty solid solution for many of them.

8:26I think, you know, their CEO has also teased an expanded toolkit on the horizon powered by their fresh funding. And he emphasized the paramount importance of addressing regulatory requirements, especially in industries like finance and insurance, because regulatory missteps can really cut back customer trust. They can inflict some very heavy penalties and also just disrupt the organization's operations overall. So Dynamo FL's privacy evaluation suite offers instant testing for data extraction weak points and generates the necessary documentation to achieve security and compliance benchmarks. So currently operating with a very tight knit team.

9:06I think there's only 17 people right now working at Dynamo FL. They're gearing up to double that. They're going to try to get to around 35 members of their team by the end of the year with this fresh round of funding that should be very feasible. I think it's a very interesting company. This is definitely a use case that is very needed today. It's going to be interesting to watch them as they grow and expand their company and see what the adoption looks like in the future.