Galtea, a startup from the Barcelona Supercomputing Center, focuses on generating realistic test scenarios for AI agents to identify failures, bias, and security risks. It has raised $3.2 million in seed funding, bringing total funding to $4.1 million. The platform automates the creation of adversarial and edge-case scenarios, allowing for efficient testing without manual input. It evaluates AI models on various metrics, including hallucination rates and bias, and has launched a self-service tier to expand access to its services.
"Galtea's core product generates test cases and synthetic user simulations from descriptions of how an AI agent is intended to behave, creating adversarial and edge-case scenarios automatically."
"The platform evaluates models across hallucination rates, bias, security vulnerabilities, and toxicity, providing structured metrics for developers and compliance teams to make informed deployment decisions."
Read at TNW | Startups-Technology
Unable to calculate read time
Collection
[
|
...
]