Dili wants to automate due diligence with AI

Stephanie Song, formerly a member of Coinbase’s business and corporate development team, was often frustrated by the volume of due diligence tasks she and her team had to complete on a daily basis.

“Analysts stay up all night working hundreds of hours doing work that no one wants to do,” Song told TechCrunch in an email interview. “At the same time, funds are deploying less capital and looking for ways to make their teams more efficient while reducing operating costs.”

Inspired to find a better way, Song teamed up with Brian Fernandez and Anand Chaturvedi, two former Coinbase colleagues, to launch Dili (not to be confused with capital of east timor), a platform that attempts to automate key investment due diligence and portfolio management steps for private equity and venture capital firms using AI.

Y Combinator graduate Dili has raised $3.6 million in venture funding to date from backers including Allianz Strategic Investments, Rebel Fund, Singularity Capital, Corenest, Decacorn, Pioneer Fund, NVO Capital, Amino Capital, Rocketship VC , Hi2 Ventures, Gaingels and Hiperempresas.

“[AI] It affects all parts of an investment fund, from analysts to partners to back-office functions,” Song said. “Fund investment professionals are looking for a differentiated edge in decision-making and can now use their wealth of data to combine their understanding of the deal with how it fits into the funds… Dili has a unique opportunity to emerge as a solution for funds in a hostile macroeconomic environment.”

Song isn’t wrong about edge-seeking funds—or any promising new way to mitigate investment risk, for that matter. V.C. reportedly They have $311 billion in unspent cash, and last year raised the lowest total ($67 billion) in seven years as they became increasingly cautious about early-stage companies.

Dili is not the first to apply AI to the due diligence process. Gartner predict that by 2025, more than 75% of executive reviews of venture capital investors and early-stage investors will be based on artificial intelligence and data analysis.

Several startups and established companies are already turning to AI to review financial documents and large amounts of data for market comparisons and reports, including Wokelo (whose clients are private equity and venture capital funds, such as Dili’s). , Ansarada, AlphaSense and Thomson Reuters (via its Clear Adverse Media unit).

But Song insists that Dili uses “first-of-its-kind” technology.

“[We can] “They offer very high precision in specific tasks, such as extracting financial metrics from large unstructured documents,” he added. “We have created custom indexing and retrieval channels tailored to specific documents to provide [our AI] models with high-quality context.”

Dili leverages GenAI, specifically large language models similar to OpenAI’s ChatGPT, to optimize investor workflows.

The platform first catalogs a fund’s historical financial data and investment decisions into a knowledge base, and then applies the aforementioned models to automate tasks such as analyzing private company databases, managing lists of due diligence requests, and searching little-known figures on the web. .

Dili recently added support for automated comparable analysis and industry benchmarking on a company’s deal portfolio. Once funds upload their transaction data, they can compare historical and current investment opportunities in one place.

“Imagine being able to receive an email with a new investment opportunity or portfolio company update and instantly have a platform that generates AI-generated deal red flags, competitive analysis, industry benchmarks, and a summary or preliminary memorandum that takes advantage of your fund’s historical investment patterns,” Song said.

The question is: can Dili’s AI (or any other AI for that matter) be trusted when it comes to managing a portfolio?

After all, AI isn’t necessarily known for sticking to the facts. fast company tried ChatGPT’s ability to summarize articles and found that the model had a tendency to get things wrong, leave out parts, and make up details not mentioned in the articles it summarized. It’s not hard to imagine how this could become a real problem in due diligence work, where accuracy is paramount.

AI can also introduce biases into the decision-making process. in an experiment carried out According to Harvard Business Review several years ago, an algorithm trained to make startup investment recommendations was found to choose white entrepreneurs over entrepreneurs of color and preferred to invest in startups with male founders. This is because the public data the algorithm was trained on reflected the fact that fewer women and founders from underrepresented groups tend to be injured in the financing process and ultimately raise less venture capital.

Then there’s the fact that some companies might not feel comfortable handling their private and sensitive data through a third-party model.

To try to allay all those fears, Song said Dili continues to fine-tune its models (many of which are open source) to reduce cases of hallucinations and improve overall accuracy. She also emphasized that private client data is not used to train Dili’s models and that Dili plans to offer a way for funds to create their own models trained with proprietary and offline fund data.

“While hedge funds and public markets have invested heavily in technology, private market data has huge untapped potential that Dili could unlock for companies,” Song said.

Dili conducted an initial pilot last year with 400 analysts and users from different types of funds and banks. But as the startup expands its team and adds new capabilities, it is looking to expand into new applications and ultimately become an “end-to-end” solution for investor due diligence and portfolio management, Song says. .

“Over time we believe this core technology we are building can be applied to all parts of the asset allocation process,” he added.