Increase Alpha Secures $3.5M to Bring Predictive AI to Hedge Funds

In a world where hedge funds chase increasingly elusive alpha, a small Washington, DC–based startup believes it has cracked the code. Increase Alpha, founded by former U.S. government AI policy advisor Sid Ghatak, has raised $3.5 million in seed funding to launch its predictive AI engine built specifically for institutional investors. The round was led by Bartt Kellermann, CEO of Battle of the Quants.

A Model Built for Markets, Not Just Math

The pitch is bold: while most hedge funds aspire to accuracy rates in the 52–55% range, Increase Alpha claims its model consistently hits 70% on hundreds of equities. Over the past three years, it has produced 90% cumulative excess returns.

Unlike the wave of generative AI experiments sweeping finance, Increase Alpha isn’t plugging ChatGPT into Bloomberg terminals. Its Predictive Artificial Intelligence (PAI) engine doesn’t rely on large language models at all. Instead, it was trained exclusively on public, compliant data, transforming corporate disclosures and market cues into validated predictions. In fact,

That specificity matters. LLMs often stumble on time-series data, transparency, and bias controls, all of which are critical in finance. Increase Alpha says its purpose-built model avoids common pitfalls such as hallucination, overfitting, or cherry-picking backtests.

“In an industry where unique, uncorrelated, and consistent alpha has become increasingly rare, we’ve demonstrated that it’s not only possible, but scalable,” said Ghatak.

In fact, the funding news follows a particularly eventful month for Increase Alpha. Independent validation came from Zanista, an AI-powered financial analytics firm, which published a research paper in SSRN titled, “Increase Alpha: Performance and Risk of an AI-Driven Trading Framework.” The report examined the company’s approach to predictive modeling and risk management, giving the startup early recognition within the quant finance community. At the same time, CEO Sid Ghatak also appeared before Congress, offering testimony on how AI can be used to improve care for U.S. veterans. Together, these milestones highlight the company’s momentum and its efforts to build credibility not only with investors, but also with policymakers and the broader AI ecosystem.

From Academia to Application

Increase Alpha’s journey began as an academic experiment at Villanova University, where Ghatak taught before entering government service. The research tested a simple premise: if the right engine could digest vast amounts of public data, it could consistently forecast stock price movements.

After eight years of academic work and four years of live trials, the results attracted hedge fund interest. Today, several top-tier firms are piloting the engine, with licensing deals in discussion.

Kellermann, who has watched the intersection of AI and trading for years, said the model’s uniqueness convinced him to invest: “This combination is so completely unique and effective that they have the potential to reimagine the entire hedge fund space.”

A Founder with Policy Credibility

Ghatak isn’t the typical fintech founder. Before launching Increase Alpha, he served as Director of the Data & Analytics Center of Excellence at the U.S. General Services Administration, co-authoring the federal AI Maturity Model. Today he acts as Chief Technical Advisor at the National Artificial Intelligence Association, advising both the White House and Congress on AI legislation.

That policy pedigree could help the startup navigate compliance concerns and build trust with institutional clients wary of black-box AI systems.

With fresh capital in hand, Increase Alpha is now focused on go-to-market execution. Hedge fund managers, Chief Investment Officers, and data buyers are already testing the signals in live environments.

The company’s pitch isn’t about replacing human fund managers, but rather giving them a new layer of predictive insight that plugs directly into existing strategies. Early pilots suggest the signals can integrate without the heavy data cleansing or infrastructure lift that typically slows adoption of alternative data.

As Ghatak put it: “Our goal isn’t to just create another model, but redefine how AI technology is applied in finance.”

If the firm can deliver on its claims, it won’t just be another fintech startup chasing AI buzzwords. It may end up shifting how hedge funds define and pursue alpha in the years ahead.