How DiDi built intelligent contact center QA with Amazon Bedrock
DiDi built a transparent, self-owned contact center quality assurance (QA) system on Amazon Bedrock, replacing an opaque third-party tool. Intent verification accuracy rose from 38% to 86%, compliance scoring topped 90%, and Voice of Customer trend analysis dropped from hours to minutes across Spanish and Portuguese support.
OpenAI's latest model, GPT-6 Astra, is now available on Amazon Bedrock for organizations to tackle their most demanding tasks. This advanced AI model enhances the depth and judgment capabilities of AI agents, enabling them to handle complex business decisions, analyze data, and automate workflows with greater efficiency.
GPT-6 Astra excels at reconciling conflicting inputs, tracing dependencies, and determining priorities. In financial analysis, it can identify discrepancies that might alter recommendations, while during contract reviews, it scrutinizes hundreds of pages within a 1 million token context window to detect high-risk provisions. For codebases, it can investigate problems across large codebases, reason through dependencies, and guide fixes from diagnosis through testing.
The model also improves computer and browser use, allowing it to work across applications and continue workflows directly through software interfaces when APIs or connectors are unavailable. It supports implicit and explicit prompt caching for workflows that reuse the same context across requests, reducing processing time and costs.
GPT-6 Astra is subject to layered security and governance for production AI, with model-level safeguards complementing Amazon Bedrock's security controls. OpenAI evaluated the model through its Preparedness Framework, reaching the highest classification for cybersecurity capability. As part of this launch, OpenAI introduced new enterprise plugins for ChatGPT Work that extend Astra's browser-use capabilities across common business applications, including data analytics, operations, and finance tools.
These plugins integrate seamlessly within existing user permissions without granting additional access.
Written by urgent.news from AWS Machine Learning's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
