Press "Enter" to skip to content

Posts tagged as “Theo AI”

Theo AI Predicts How Much It Would Take to Settle a Lawsuit

A set of weight scales on a table, depicting justice

Theo AI Predicts How Much It Would Take to Settle a Lawsuit

By Deborah Yao on December 3, 2025

CEO Patrick Ip says the system has an accuracy rate of 85% when backtested against past cases.

Continue reading

Why Insurers Need AI that Understands Risk, Not Just Data
Why Insurers Need AI that Understands Risk, Not Just Data
By Srinath Chandramohan
An opinion piece by EY's insurtech consulting leader and our columnist
AI Tool Review: ChatGPT Enterprise as Analytics Copilot
AI Tool Review: ChatGPT Enterprise as Analytics Copilot
By Sandeep Mahajan
High leverage for analytics teams, but only with engineered governance guardrails, writes our reviewer and columnist.
The Rise of Delegated Commerce: When AI Starts Spending for You
The Rise of Delegated Commerce: When AI Starts Spending for You
By Arjun Wadwalkar
When people let AI make purchases for them, it's not due to novelty. It's exhaustion.
Why Isolated Legal Tools Are the Biggest Liability in the Age of AI
Why Isolated Legal Tools Are the Biggest Liability in the Age of AI
By Sarah Webb
An opinion piece by the chief operating officer of LawVu

Training LLMs Not to Lie
By OpenAI researchers on January 23, 2026
Large language models might be dishonest about their covert actions. OpenAI researchers propose a self-confession approach as a solution.
DeepSeek Proposes New Method to Improve Stability in LLMs
By DeepSeek researchers on January 6, 2026
MHC lets large language models train more reliably at scale without major increases in compute cost.
Small Language Models are the Future of Agentic AI
By Researchers from Nvidia and Georgia Institute of Technology on December 2, 2025
SLMs are inherently more suitable for agentic systems, according to their paper.