Since the launch of Amazon Nova at AWS re:Invent 2024, we have seen adoption trends across industries, with notable gains in operational efficiency, compliance, and customer satisfaction. With its capabilities in secure, multimodal AI and domain customization, Nova is enhancing workflows and enabling cost efficiencies across core use cases.
In this post, we share four high-impact, widely adopted use cases built with Nova in Amazon Bedrock, supported by real-world customers deployments, offerings available from AWS partners, and experiences. These examples are ideal for organizations researching their own AI adoption strategies and use cases across industries.
Customer service
Traditional chatbots often frustrate users with scripted, inflexible responses that fail to understand context or intent. For enterprises, these are missed opportunities to resolve issues quickly, lower support costs, and drive customer loyalty. AI-powered applications can understand natural language, adapt to individual customer needs, and integrate with backend systems in real time. Organizations are transforming support from a cost center into a strategic driver of satisfaction and retention. These are often high-volume and interactive scenarios, so the balance of cost, speed, and intelligence is critical.
Customer service applications built with Nova in Amazon Bedrock can seamlessly integrate with business data stored with AWS, and offer the security, privacy, and reliability for production use in enterprise environments.
- Infosys, a leading global IT services and consulting organization, developed Infosys Event AI for real-time transcription, multilingual translation, and intelligent summarization of live event content. Infosys Event AI is built with Amazon Nova Pro in Amazon Bedrock. During a recent event in Bangalore, the AI assistant handled around 230 users per minute and was queried an average of 57 times per minute, generating more than 9,000 session summaries. This solution enhanced knowledge retention, engagement, and inclusivity by making event insights instantly accessible in multiple languages and formats for hearing-impaired and remote participants. By transforming event content into a persistent, searchable multilingual knowledge asset, Infosys Event AI accelerates learning and collaboration.
- Fortinet, an AWS Partner and cybersecurity company, uses Amazon Nova Micro to power its AI support assistant, delivering significant performance improvements at a fraction of the cost. By switching to Nova Micro in Amazon Bedrock, Fortinet achieved an 85 times reduction in inference costs, dramatically lowering TCO while maintaining rapid response times. The assistant now helps users quickly navigate complex documentation across more than 60 products, improving support efficiency and elevating customer satisfaction.
- Amazon Customer Service uses Nova with its AI-driven issue resolution system. The system is a two-step approach combining intent detection and issue resolution. Amazon Customer Service customized Nova Micro, resulting in 76.9% accuracy for in-domain issues and 69.2% in generalization testing, surpassing current baselines by 5.4% and 7.3%, respectively. Additionally, Nova Lite is used for tool selection, achieving 86.1% accuracy and 4.8% improvement over existing systems.
- AWS Summit New York City 2025 was attended by 18,000 participants, featuring the AI assistant Diana for customer service developed with Nova Sonic. By dialing a phone number, the Sonic-powered voice assistant answered hundreds of queries about the event, including session details, location, and FAQs.
Search
Large enterprises face slow, siloed, and inefficient search across vast stores of structured and unstructured data, costing time, productivity, and customer responsiveness. By adopting AI-powered, multimodal search that understands natural language and enforces secure access, organizations can deliver instant, relevant answers from documents, images, and technical files. This accelerates decision-making, shortens deal cycles, improves customer satisfaction, and reduces the cost of knowledge discovery at scale. Search applications increasingly rely on a mix of information across modalities, including text, documents, images, and video.
Nova is among the fastest and most cost-effective multimodal models, offering vision fine-tuning capabilities. Nova also integrates with broader Amazon models including Amazon Titan Multimodal Embeddings and data services including Amazon OpenSearch Service for more robust search capabilities and performance.
- Siemens faced growing performance bottlenecks as its massive datasets strained traditional search systems, slowing retrieval speeds and impacting productivity across its global operations. To address this, Siemens integrated Amazon Nova, achieving a threefold boost in search performance that dramatically accelerated data retrieval and improved workflow efficiency. Amazon Nova delivers high-speed, scalable search capabilities, and Siemens’s implementation facilitates seamless integration with existing systems, maintaining business continuity with minimal disruption. This enhanced user experience and positioned Siemens to handle future data growth with ease, supported by continuous performance monitoring and tight infrastructure alignment.
- CBRE Global Pulse System (GPS)—built with Amazon Nova Pro in Amazon Bedrock and OpenSearch Service—transforms property search across thousands of users worldwide. Built in partnership with AWS Professional Services and GenAI Specialists, GPS replaces slow, fragmented legacy systems with an AI-driven, multimodal search platform capable of handling complex queries, massive PDFs, and strict permission controls. Key results include 75% faster document ingestion, 70% lower database latency, 87% faster keyword searches, and 51% faster natural language queries. When fully deployed to over 6,000 users later in 2025, GPS is projected to save over 98,000 employee workdays annually, unlocking $320,000 ARR and significant operational efficiency. By shifting from Anthropic’s Claude Sonnet to Nova Pro and Anthropic’s Claude Haiku 3, CBRE also cut AI inference costs by 3.5 times and 12 times, respectively, without sacrificing accuracy.
Video understanding and analysis
Organizations are adopting video understanding applications to drive business value across multiple fronts, including customer behavior analysis, traffic patterns, and manufacturing quality control. Security and safety benefits are realized through real-time threat detection and workplace safety monitoring, and customer experience is enhanced through personalized content recommendations and improved content searchability. Organizations gain competitive advantage through data-driven decision-making and innovation in service delivery, while reducing costs by minimizing manual review processes and decreasing security incidents. This comprehensive approach to video analysis helps companies extract insights from their video data, ultimately leading to improved operations, better decision-making, and enhanced customer experiences. As developers build, iterate, and evolve these applications, there is a growing demand to natively understand video as opposed to dealing with the overhead of frames, time stamps, and synchronization.
Amazon Nova models can analyze, classify, and summarize information in the video based on provided instructions. Applications built with Nova understanding models in Amazon Bedrock offer comprehensive analysis of multiple video formats through flexible input methods, with the ability to analyze, classify, and summarize video content while handling files up to 1 GB through Amazon Simple Storage Service (Amazon S3) integration.
- Bitcentral partnered with Caylent to transform how archived content is discovered, accessed, and reused. Using Nova Pro in Amazon Bedrock, Caylent deployed a solution that aligned with the needs of journalists, producers, and broadcasters across more than 1,600 client sites. By embedding semantic video search, contextual metadata generation, and AI-powered content analysis into its workflows, Bitcentral redefined how archived footage is indexed, discovered, and reused. Journalists and producers can now surface high-value content in real time and unlock new revenue streams.
- Loka, an AWS Premier Partner, built a video surveillance offering to automatically identify and classify millions of visual events in video footage. This system effectively distinguishes between routine events and critical incidents, helping filter out non-essential activities and alerts. The solution proved highly successful, reducing irrelevant alerts by 55% while maintaining a threat detection rate above 97%. By implementing this automated filtering system, Loka doubled video monitoring efficiency for their client. The tool, built on Amazon Bedrock using Amazon Nova Pro, significantly reduced the workload for human operators while improving overall threat detection capabilities.
- Accenture Spotlight can analyze long-form videos and automatically generate personalized short-form clips and highlights, which are particularly useful for sports content like soccer, Formula 1, and rugby. Spotlight is capable of matching content to specific audience demographics and can process real-time CCTV footage in retail settings to create personalized offers. The system is built with Amazon Nova in Amazon Bedrock and operates through three specialized super agents working under a central orchestrator. Spotlight can process videos in minutes rather than the traditional hours or days, while achieving cost savings that are 10 times better than conventional methods. The solution is versatile enough to be used across different industries, from media and entertainment to retail, while maintaining high quality standards and brand alignment through its human-in-the-loop quality assurance option.
Creative content generation
Organizations are seeking ways to revolutionize creative content generation including stock imagery, marketing campaign assets, and product visualizations. It is often slowed down by fragmented workflows, high production costs, and the need to continuously balance scale with personalization. Marketing teams struggle to keep up with the demand for fresh, high-quality assets across multiple channels, while creative fatigue and long lead times limit their agility.
Amazon Nova addresses these challenges with Nova Canvas and Nova Reel: high-quality creative models that transform text and image inputs into professional-grade images and videos. Nova creative models are designed to deliver customizable visual content with control features, making creative content generation accessible and efficient for media, entertainment, retail, marketing, and advertising industries.
- Dentsu is reimagining how ads come to life with Amazon Nova creative generation models. What used to take weeks of brainstorming, filming, and editing now happens in days. Their creative teams can sketch out an idea in plain language and watch it turn into polished videos and custom images, ready for markets across the globe in over 200 languages. Built-in safeguards like moderation, watermarking, and IP indemnity mean every piece stays brand safe. For Co-op, Dentsu went a step further—pairing Nova with Amazon Ads to design custom audience profiles that delivered a +4-point lift in brand preference among 25–34-year-olds and a +5-point lift in favorability among affluent shoppers.
- Quantiphi, an AWS Premier Global Consulting Partner, developed Qreator, a generative AI-powered marketing content creation service built on AWS. Their service helps marketers create content through natural language prompts while maintaining brand consistency and cross-channel adaptability. With Qreator, business can achieve an approximate 30% reduction in content creation time and get to market approximately 40% faster, automating what was a manual process, and improving consistency across formats and channels.
- The Fragrance Lab is a unique AWS activation that was showcased at the Cannes Lions International Festival of Creativity. It demonstrates how to build personalized products and campaign assets using Amazon Nova foundation models in Amazon Bedrock. Although our activation at Cannes Lions focused on personalized fragrance development and ad campaign creation, the underlying architecture and methodology can be adapted across diverse categories, such as fashion, food, and beverage, opening endless possibilities for customized customer experiences. The Fragrance Lab activation won two International Business Awards: Gold for Exhibition Event Experience and Silver for Experiential Event.
Conclusion
The four use cases presented in this post demonstrate the utility of Amazon Nova across industries and applications. From Infosys’s Event AI improving accessibility and engagement, to CBRE’s revolutionary property search system, to Loka’s intelligent video surveillance, and Dentsu’s creative content generation, each implementation showcases significant, measurable improvements in efficiency, cost reduction, and customer satisfaction.
Organizations using Amazon Nova are achieving tangible business outcomes through evidence-based adoption strategies. By partnering with Amazon and AWS Partners, organizations are accelerating their AI transformation while maintaining strong foundations in security, compliance, and privacy-by-design principles.
To get started building with Nova, visit the Amazon Nova user guide or visit the AWS console.
About the Authors
Abhinav Bhargava is a Sr Product Marketing Manager at AWS on the Amazon Nova team, where he focuses on scaling generative AI adoption through customer-centric solutions. With a background in design and sustainability, he brings a unique perspective to connecting technology and creativity to drive enterprise innovation. Based in Seattle, Abhinav enjoys playing volleyball, traveling, and learning about new cultures.
Raechel Frick is a Sr Product Marketing Manager at AWS. With over 20 years of experience in the tech industry, she brings a customer-first approach and growth mindset to building integrated marketing programs. Based in the greater Seattle area, Raechel balances her professional life with being a soccer mom and after-school carpool manager, demonstrating her ability to excel both in the corporate world and family life.