5 Ways to Future-Proof Your Research Process for Collapsing AI Costs

1. Make your investment thesis the centre of everything.

Every research artifact – meeting notes, models, ESG scorecards – should tie back to the thesis it supports. This lets AI continuously re-evaluate: “Does this thesis still hold given new evidence?”

2. Treat AI providers as interchangeable plug-ins.

Keep your investment process, thesis structures, and portfolio positions in your own system. OpenAI, Anthropic, Gemini – swap models as prices and capabilities shift. Never lock yourself to one vendor.

3. Build for information overload, not scarcity.

When analysis is cheap, attention becomes the bottleneck. Design filters, alerts, and dashboards that surface only where evidence diverges most from your current thesis.

4. Double down on proprietary data.

Everyone will have access to the same frontier models running on the same public data. No edge there. Your edge comes from proprietary research notes, company meetings, engagement history, and how deeply AI integrates with that data.

5. Plan for regular AI upgrades.

Treat model upgrades like you treat index or pricing feed updates. Your systems should adopt new models, bigger context windows, and better tools without breaking workflows.

—