Normalized claim
I don't know responses reduction: 87% decrease
"87 percent decrease in 'I don't know' chatbot responses"
inGenious.ai Improves Chatbot Comprehension by 80 percent with Amazon Nova and Amazon Bedrock
Use case typeCustomer service automationUpdated Jun 13, 2026
inGenious.ai specializes in the development of artificial intelligence (AI)-powered chatbots for non-technical users and businesses that need conversational interfaces integrated with multiple business systems. The company built a next-generation generative AI chatbot on Amazon Bedrock and selected Amazon Nova Micro after testing multiple LLMs through a single interface. Amazon Bedrock Knowledge Bases were used to connect the chatbot to private customer data sources for secure RAG, and Amazon Bedrock Guardrails were used to protect sensitive information and support compliance requirements. The solution was deployed into production in eight weeks and used to rewrite/add context and generate summaries for smoother agent handovers.
- Organization
- inGenious.ai
- Industry
- Finance
- Location
- Australia
- Published
- June 2026
Reported outcomes
10,000 queries
Queries answered correctly on first attemptOther quantified impact
−87%I don't know responses reduction300 millisecondsIntent classification latency14 pointsNet Promoter Score increase−24%Agent handover rate reduction−33%Support ticket volume reduction
Strategic outcomes
Customer experience & trustImproved handling of complex queriesCustomer experience & trustReduced unsupported chatbot responsesSpeed & agilityEnabled real-time low-latency responsesCustomer experience & trustReduced agent handovers and support tickets
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Queries answered correctly on first attempt: 10,000 queries increase
"enabled approximately 10,000 more queries to be answered correctly on the first attempt"
Normalized claim
Intent classification latency: 300 milliseconds decrease
"sub-300 millisecond latency for real-time intent classification"
Normalized claim
Net Promoter Score increase: 14 points increase
"a 14-point increase in its Net Promoter Score"
Normalized claim
Agent handover rate reduction: 24% decrease
"reduce agent handover rates by 24 percent"
Normalized claim
Support ticket volume reduction: 33% decrease
"support ticket volume dropped by 33 percent"
Normalized claim
Long and complex query handling improvement: 80% increase
"an 80 percent improvement in handling long and complex queries"
Normalized claim
Comprehension failures reduction: 28% decrease
"a 28 percent decrease in comprehension failures"
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- inGenious.ai, National Australia Bank, Latitude Financial Services
- Provider
- AWS
- Maturity
- Production
- Linked source
- AWS Solutions Case Study
The solution was deployed into production in eight weeks and used to rewrite/add context and generate summaries for smoother agent handovers
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 3 of 3
- 1Customer Service Automation
- 2Conversational AI
- 3RAG
- Traditional chatbots struggled with lengthy, complex queries and poor syntax.
- The company needed subsecond responses and stronger comprehension without sacrificing security and compliance for financial services customers.
- The solution had to align with SOC 2 and APRA standards while avoiding heavy custom integration overhead.
- Built production generative AI chatbots on Amazon Bedrock.
- Tested multiple LLMs through a single interface and chose Amazon Nova Micro for subsecond performance and instruction following.
- Used Amazon Bedrock Knowledge Bases to deploy secure, scalable retrieval-augmented generation against customer data sources.
- Applied Amazon Bedrock Guardrails to protect sensitive information and keep the architecture within AWS.
- Used the chatbot to rewrite and add context to customer queries and to generate summaries for downstream agent handovers.
- At National Australia Bank, the chatbot reduced "I don't know" responses by 87% and enabled about 10,000 more queries to be answered correctly on the first attempt.
- NAB also achieved sub-300 millisecond latency for real-time intent classification and a 14-point increase in Net Promoter Score.
- At Latitude Financial Services, agent handover rates fell by 24% and support ticket volume dropped by 33%.
- The updated platform delivered an 80% improvement in handling long and complex queries, a 28% decrease in comprehension failures, and a 31% gain in overall accuracy.
Architecture
inGenious.ai built its chatbot on Amazon Bedrock, evaluated multiple foundation models through a single integration, selected Amazon Nova Micro, used Amazon Bedrock Knowledge Bases for secure RAG, and used Amazon Bedrock Guardrails for sensitive-data protection and compliance-oriented controls.
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jun 8, 2026Publisher: AWSEvidence: VendorConfidence: Medium
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
AI patterns
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.