AI agents currently face reliability challenges, with benchmarks like LiveBench 2.0 showing success rates around 75-83% for complex tasks. EON's tests in simulated corporate environments also found accuracy rates around 77% for the most successful models. These figures indicate that agents often require supervision and struggle with unpredictable environments.
Plurai, a new startup, aims to improve AI agent performance using simulation techniques previously applied in self-driving car development. Co-founders Ilan Kadar and Elad Levi, who have backgrounds in autonomous driving technology, observed similar reliability patterns in AI agents as in early self-driving systems. Their solution involves creating a simulation layer to prepare agents for difficult scenarios.
Plurai announced $13 million in seed funding, led by Team8, with participation from U&I, Velocity, and Mercer. The company's platform creates a digital twin of a company's electronic systems, including knowledge bases and tools accessible to an agent. It then maps possible agent flows and uses this graph to evaluate and fix potential edge cases, working with existing agentic tools like Claude Code, Codex, Copilot, and Cursor.
The simulation approach is designed to address issues where agents, like probabilistic systems in unpredictable environments, might fail. Examples include agents ignoring spending policies when booking travel or struggling to combine information from multiple datasets. By pre-training agents with synthetic data from these simulations, Plurai seeks to make them more reliable and cost-effective for real-world applications.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's briefNew every morning, and the back catalogue is archived by date.
Plurai secured $13 million in seed funding to develop a simulation platform for AI agents. The platform uses synthetic data and digital twins of company systems to train agents, addressing reliability issues observed in current AI agent benchmarks. This approach aims to prepare models for complex edge cases, similar to methods used in autonomous driving development.