Franklin Templeton develops investment research workflows with Codex
Franklin Templeton is developing governed workflows that use Codex, agents, and a semantic layer to help portfolio managers compare research viewpoints and examine investment risks.
AT A GLANCE
Key points
- Codex is one of several tools used in connected research workflows.
- Agents and a semantic layer connect data relationships with task context.
- Research, risk, compliance, and trading perspectives are compared within the same workflow.
- Production rollout includes security review, data controls, and human judgment.
What is changing?
According to OpenAI’s case study published on November 5, Franklin Templeton is starting with equity research to develop AI workflows that help portfolio managers compare different views and examine risks. Codex is one of the tools used in the system. In early experiments, models could connect to the database but failed to understand relationships across the data. Adding agents and a semantic layer brought content from trusted connectors, portfolio context, and task steps into a shared connected-intelligence layer.
Planned role-specific workflows combine perspectives from research, risk, compliance, and trading teams. Hackathons and an Investment AI Champions network move prototypes toward production with security review, expert support, and data controls. The goal is not to automate portfolio decisions, but to surface diverse research views and potential risks. Final judgment and accountability remain with human experts.
Explore the original sources
Source date: 5 November 2026.
Published by AI Radar: 6 Nov 2026, 15:02 (Istanbul, UTC+3). Updated: 6 Nov 2026, 16:07.
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…