• The Shift Brief
  • Posts
  • Wall Street’s “New” AI Strategy: Build, Don’t Buy

Wall Street’s “New” AI Strategy: Build, Don’t Buy

The Shift Brief | Week of July 28th

Summary:

A recent article from The Next Platform (available here) highlights a growing shift among financial institutions: building internal infrastructure to train proprietary AI models. Rather than relying on commercial LLMs, firms are prioritizing control, customization, and security.

From hedge funds to large banks, leaders are recognizing that off-the-shelf models, such as those from OpenAI or Anthropic, often fall short in terms of precision, compliance, and context. They're now developing dedicated AI pipelines that combine high-performance compute, domain-specific data, and custom model architectures.

Key drivers include:

Data sensitivity: Firms want to ensure proprietary insights stay internal

Model alignment: General-purpose models often misunderstand financial nuance

Performance: Tailored models trained on internal data consistently outperform generic tools when the stakes are high

Why It Matters:

This marks a shift in how financial firms think about AI. It's no longer about plugging into what’s available, but about owning the stack and shaping it around your firm's DNA.

The question for many teams now is simple:

What part of your research, data, or IP is too valuable to trust to a public model?

Where We Fit In:

Shift is built to serve as the secure connective layer between your internal tools, systems, and data. Our platform gives you the control and flexibility to create AI-enhanced research workflows that feel custom, without needing to build a full model infrastructure from scratch.

~Ryan Erickson, Founding Executive

TLDR: What you need to know about AI in Finance

Prompt of the Week

Use this to sharpen your thinking before making a macro or sector call:

Act as a skeptical investment committee member. Challenge the bullish case for [Sector or Theme], pointing out key risks, historical parallels, and counterarguments a seasoned PM might raise.

💡Great for pressure-testing your thesis or preparing for an internal debate, especially when everyone seems to agree on the direction. Perfect for analysts writing investment memos or PMs reviewing portfolio tilts before rebalancing.

BASIS PLATFORM: What Happens Behind the Scenes

Basis connects to your emails, research notes, financial models, internal conversations, and other tools. It classifies the content, tags it with relevant context (like tickers or topics), and stores it securely.

From there, it pulls everything into a single workspace, uses AI to surface key developments, and automates workflows to keep your team aligned and moving fast.

It’s like giving your investment team a command center: one view, smarter insights, less chasing.

About Shift
Shift builds AI-powered research tools that help investment teams surface insights and drive better decisions. We're a team of builders and finance experts based in Charlottesville, VA. Learn more at shifthq.ai.