{"id":128169,"date":"2025-11-08T05:32:12","date_gmt":"2025-11-08T05:32:12","guid":{"rendered":"https:\/\/www.newsbeep.com\/ie\/128169\/"},"modified":"2025-11-08T05:32:12","modified_gmt":"2025-11-08T05:32:12","slug":"democratizing-ai-how-thomson-reuters-open-arena-supports-no-code-ai-for-every-professional-with-amazon-bedrock","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ie\/128169\/","title":{"rendered":"Democratizing AI: How Thomson Reuters Open Arena supports no-code AI for every professional with Amazon Bedrock"},"content":{"rendered":"\n

This post is cowritten by Laura Skylaki, Vaibhav Goswami, Ramdev Wudali and Sahar El Khoury from Thomson Reuters.<\/p>\n

Thomson Reuters (TR)<\/a> is a leading AI and technology company dedicated to delivering trusted content and workflow automation solutions. With over 150 years of expertise, TR provides essential solutions across legal, tax, accounting, risk, trade, and media sectors in a fast-evolving world.<\/p>\n

TR recognized early that AI adoption would fundamentally transform professional work. According to TR\u2019s 2025 Future of Professionals Report,<\/a> 80% of professionals anticipate AI significantly impacting their work within five years, with projected productivity gains of up to 12 hours per week by 2029. To unlock this immense potential, TR needed a solution to democratize AI creation across its organization.<\/p>\n

In this blog post, we explore how TR addressed key business use cases with Open Arena, a highly scalable and flexible no-code AI solution powered by Amazon Bedrock<\/a> and other AWS services such as Amazon OpenSearch Service<\/a>, Amazon Simple Storage Service (Amazon S3)<\/a>, Amazon DynamoDB<\/a>, and AWS Lambda<\/a>. We\u2019ll explain how TR used AWS services to build this solution, including how the architecture was designed, the use cases it solves, and the business profiles that use it. The system demonstrates TR\u2019s successful approach of using existing TR services for rapid launches while supporting thousands of users, showcasing how organizations can democratize AI access and support business profiles (for example, AI explorers and SMEs) to create applications without coding expertise.<\/p>\n

Introducing Open Arena: No-code AI for all <\/p>\n

TR introduced Open Arena<\/a> to non-technical professionals to create their own customized AI solutions. With Open Arena users can use cutting-edge AI powered by Amazon Bedrock in a no-code environment, exemplifying TR\u2019s commitment to democratizing AI access.<\/p>\n

Today, Open Arena supports:<\/p>\n

High adoption: ~70% employee adoption, with\u00a019,000 monthly active users.
\n Custom solutions: Thousands of customized AI solutions created\u00a0without coding, used for internal workflows or integrated into TR products for customers.
\n Self-served functionality: 100% self-served functionality, so that users, irrespective of technical background, can develop, evaluate, and deploy generative AI solutions. <\/p>\n

The Open Arena journey: From prototype to enterprise solution <\/p>\n

Conceived as a rapid prototype, Open Arena was developed in under six weeks at the onset of the generative AI boom in early 2023 by TR Labs \u2013 TR\u2019s dedicated applied research division focused on the research, development, and application of AI and emerging trends in technologies. The goal was to support internal team exploration of large language models (LLMs) and discover unique use cases by merging LLM capabilities with TR company data.<\/p>\n

Open Arena\u2019s introduction significantly increased AI awareness, fostered developer-SME collaboration for groundbreaking concepts, and accelerated AI capability development for TR products. The rapid success and demand for new features quickly highlighted Open Arena\u2019s potential for AI democratization, so TR developed an enterprise version of Open Arena. Built on the\u00a0TR AI Platform<\/a>, Open Arena enterprise version offers secure, scalable, and standardized services covering the entire AI development lifecycle, significantly accelerating time to production.<\/p>\n

The Open Arena enterprise version uses existing system capabilities for enhanced data access controls, standardized service access, and compliance with TR\u2019s governance and ethical standards. This version introduced self-served capabilities so that every user, irrespective of their technical ability, can create, evaluate, and deploy customized AI solutions in a no-code environment.<\/p>\n

\u201cThe foundation of the AI Platform has always been about empowerment; in the early days it was about empowering Data Scientists but with the rise of Gen AI, the platform adapted and evolved on empowering users of any background to leverage and create AI Solutions.\u201d<\/p>\n

\u2013 Maria Apazoglou, Head of AI Engineering, CoCounsel<\/p>\n

As of July 2025, the TR Enterprise AI Platform consists of 15 services spanning the entire AI development lifecycle and user personas. Open Arena remains one of its most popular, serving 19,000 users each month, with increasing monthly usage.<\/p>\n

Addressing key enterprise AI challenges across user types <\/p>\n

Using the TR Enterprise AI Platform, Open Arena helped thousands of professionals transition into using generative AI. AI-powered innovation is now readily in the hands of everyone, not just AI scientists.<\/p>\n

Open Arena successfully addresses four critical enterprise AI challenges:<\/p>\n

Enablement:\u00a0Delivers AI solution building with consistent LLM and service provider experience and support for various user personas, including non-technical.
\n Security and quality:\u00a0Streamlines AI solution quality tracking using evaluation and monitoring services, whilst complying with data governance and ethics policies.
\n Speed and reusability:\u00a0Automates workflows and uses existing AI solutions and prompts.
\n Resources and cost management:\u00a0Tracks and displays generative AI solution resource consumption, supporting transparency and efficiency. <\/p>\n

The solution currently supports several AI experiences, including tech support, content creation, coding assistance, data extraction and analysis, proof reading, project management, content summarization, personal development, translation, and problem solving, catering to different user needs across the organization.<\/p>\n

\"Different<\/p>\n

\"Different<\/p>\n

Figure 1. Examples of Open Arena use cases.<\/p>\n

AI explorers use Open Arena to speed up day-to-day tasks, such as summarizing documents, engaging in LLM chat, building custom workflows, and comparing AI models. AI creators and Subject Matter Experts (SMEs) use Open Arena to build custom AI workflows and experiences and to evaluate solutions without requiring coding knowledge. Meanwhile, developers can develop and deploy new AI solutions at speed, training models, creating new AI skills, and deploying AI capabilities.<\/p>\n

Why Thomson Reuters selected AWS for Open Arena <\/p>\n

TR strategically chose AWS as a primary cloud provider for Open Arena based on several critical factors:<\/p>\n

Comprehensive AI\/ML capabilities:\u00a0Amazon Bedrock offers easy access to a choice of high-performing foundation models from leading AI companies like AI21 Labs, Anthropic, Cohere, DeepSeek, Luma AI, Meta, Mistral AI, OpenAI, Qwen, Stability AI, TwelveLabs, Writer, and Amazon. It supports simple chat and complex RAG workflows, and integrates seamlessly with TR\u2019s existing Enterprise AI Platform.
\n Enterprise-grade security and governance:\u00a0Advanced security controls, model access using RBAC, data handling with enhanced security features, single sign-on (SSO) enabled, and clear operational and user data separation across AWS accounts.
\n Scalable infrastructure:\u00a0Serverless architecture for automatic scaling, pay-per-use pricing for cost optimization, and global availability with low latency.
\n Existing relationship and expertise:\u00a0Strong, established relationship between TR and AWS, existing Enterprise AI Platform on AWS, and deep AWS expertise within TR\u2019s technical teams. <\/p>\n

\u201cOur long-standing partnership with AWS and their robust, flexible and innovative services made them the natural choice to power Open Arena and accelerate our AI initiatives.\u201d <\/p>\n

\u2013 Maria Apazoglou, Head of AI Engineering, CoCounsel<\/p>\n

Open Arena architecture: Scalability, extensibility, and security <\/p>\n

Designed for a broad enterprise audience, Open Arena prioritizes scalability, extensibility and security while maintaining simplicity for non-technical users to create and deploy AI solutions. The following diagram illustrates the architecture of Open Arena.<\/p>\n

\"Architecture<\/p>\n

Figure 2. Architecture design of Open Arena.<\/p>\n

The architecture design facilitates enterprise-grade performance with clear separation between capability and usage, aligning with TR\u2019s enterprise cost and usage tracking requirements.<\/p>\n

The following are key components of the solution architecture:<\/p>\n

No-code interface:\u00a0Intuitive UI, visual workflow builder, pre-built templates, drag-and-drop functionality.
\n Enterprise integration:\u00a0Seamless integration with TR\u2019s Enterprise AI Platform, SSO enabled, data handling with enhanced security, clear data separation.
\n Solution management:\u00a0Searchable repository, public\/private sharing, version control, usage analytics. <\/p>\n

TR developed Open Arena using AWS services such as Amazon Bedrock, Amazon OpenSearch, Amazon DynamoDB, Amazon API Gateway, AWS Lambda, and AWS Step Functions. It uses Amazon Bedrock for foundational model interactions, supporting simple chat and complex Retrieval-Augmented Generation (RAG) tasks. Open Arena uses Amazon Bedrock Flows as the custom workflow builder where users can drag-and-drop components like prompts, agents, knowledge bases and Lambda functions to create sophisticated AI workflows without coding. The system also integrates with AWS OpenSearch for knowledge bases and external APIs for advanced agent capabilities.<\/p>\n

For data separation, orchestration is managed using the Enterprise AI Platform AWS account, capturing operational data. Flow instances and user-specific data reside in the user\u2019s dedicated AWS account, stored in a database. Each user\u2019s data and workflow executions are isolated within their respective AWS accounts, which is required for complying with Thomson Reuters data sovereignty and enterprise security policies with strict regional controls. The system integrates with Thomson Reuters SSO solution to automatically identify users and grant secure, private access to foundational models.<\/p>\n

The orchestration layer, centrally hosted within the Enterprise AI Platform AWS account, manages AI workflow activities, including scheduling, deployment, resource provisioning, and governance across user environments.<\/p>\n

The system features fully automated provisioning of\u00a0 Amazon Bedrock Flows directly within each user\u2019s AWS account, avoiding manual setup and accelerating time to value. Using AWS Lambda for serverless compute and DynamoDB for scalable, low-latency storage, the system dynamically allocates resources based on real-time demand. This architecture makes sure prompt flows and supporting infrastructure are deployed and scaled to match workload fluctuations, optimizing performance, cost, and user experience.<\/p>\n

\u201cOur decision to adopt a cross-account architecture was driven by a commitment to enterprise security and operational excellence. By isolating orchestration from execution, we make sure that each user\u2019s data remains private and secure within their own AWS account, while still delivering a seamless, centrally-managed experience. This design empowers organizations to innovate rapidly without compromising compliance or control.\u201d<\/p>\n

\u2013 Thomson Reuters\u2019 architecture team<\/p>\n

Evolution of Open Arena: From classic to Amazon Bedrock Flows-powered chain builder <\/p>\n

Open Arena has evolved to cater to varying levels of user sophistication:<\/p>\n

Open Arena v1 (Classic):\u00a0Features a form-based interface for simple prompt customization and basic AI workflow deployment within a single AWS account. Its simplicity appeals to novice users for straightforward use cases, though with limited advanced capabilities.
\n Open Arena v2 (Chain Builder):\u00a0Introduces a robust, visual workflow builder interface, enabling users to design complex, multi-step AI workflows using drag-and-drop components. With support for advanced node types, parallel execution, and seamless cross-account deployment, Chain Builder dramatically expands the system\u2019s capabilities and accessibility for non-technical users. <\/p>\n

Thomson Reuters uses Amazon Bedrock Flows as a core feature of Chain Builder. Users can define, customize, and deploy AI-driven workflows using Amazon Bedrock models. Bedrock Flows supports advanced workflows combining multiple prompt nodes, incorporating AWS Lambda functions, and supporting sophisticated RAG pipelines. Operating seamlessly across user AWS accounts, Bedrock Flows facilitates secure, scalable execution of personalized AI solutions, serving as the fundamental engine for the Chain Builder workflows and driving TR\u2019s ability to deliver robust, enterprise-grade automation and innovation.<\/p>\n

What\u2019s next? <\/p>\n

TR continues to expand Open Arena\u2019s capabilities through the strategic partnership with AWS, focusing on:<\/p>\n

Driving further adoption of Open Arena\u2019s DIY capabilities.
\n Enhancing flexibility for workflow creation in Chain Builder with custom components, such as inline scripts.
\n Developing new templates to represent common tasks and workflows.
\n Enhancing collaboration features within Open Arena.
\n Extending multimodal capabilities and model integration.
\n Expanding into new use cases across the enterprise. <\/p>\n

\u201cFrom innovating new product ideas to reimagining daily tasks for Thomson Reuters employees, we continue to push the boundaries of what\u2019s possible with Open Arena.\u201d<\/p>\n

\u2013 Maria Apazoglou, Head of AI Engineering, CoCounsel<\/p>\n

Conclusion <\/p>\n

In this blog post, we explored how Thomson Reuters\u2019 Open Arena demonstrates the successful democratization of AI across an enterprise by using AWS services, particularly Amazon Bedrock and Bedrock Flows. With 19,000 monthly active users and 70% employee adoption, the system proves that no-code AI solutions can deliver enterprise-scale impact while maintaining security and governance standards.<\/p>\n

By combining the robust infrastructure of AWS with innovative architecture design, TR has created a blueprint for AI democratization that empowers professionals across technical skill levels to harness generative AI for their daily work.<\/p>\n

As Open Arena continues to evolve, it exemplifies how strategic cloud partnerships can accelerate AI adoption and transform how organizations approach innovation with generative AI.<\/p>\n

About the authors <\/p>\n

\"\"Laura Skylaki, PhD, leads the Enterprise AI Platform at Thomson Reuters, driving the development of GenAI services that accelerate the creation, testing and deployment of AI solutions, enhancing product value. A recognized expert with a doctorate in stem cell bioinformatics, her extensive experience in AI research and practical application spans legal, tax, and biotech domains. Her machine learning work is published in leading academic journals, and she is a frequent speaker on AI and machine learning<\/p>\n

\"\"Vaibhav Goswami is a Lead Software Engineer on the AI Platform team at Thomson Reuters, where he leads the development of the Generative AI Platform that empowers users to build and deploy generative AI solutions at scale. With expertise in building production-grade AI systems, he focuses on creating tools and infrastructure that democratize access to cutting-edge AI capabilities across the enterprise.<\/p>\n

\"\"Ramdev Wudali is a Distinguished Engineer, helping architect and build the AI\/ML Platform to enable the Enterprise user, data scientists and researchers to develop Generative AI and machine learning solutions by democratizing access to tools and LLMs. In his spare time, he loves to fold paper to create origami tessellations, and wearing irreverent T-shirts<\/p>\n

\"\"As the director of AI Platform Adoption and Training, Sahar El Khoury guides users to seamlessly onboard and successfully use the platform services, drawing on her experience in AI and data analysis across robotics (PhD), financial markets, and media.<\/p>\n

\"\"Vu San Ha Huynh is a Solutions Architect at AWS with a PhD in Computer Science. He helps large Enterprise customers drive innovation across different domains with a focus on AI\/ML and Generative AI solutions.<\/p>\n

\"\"Paul Wright is a Senior Technical Account Manager, with over 20 years experience in the IT industry and over 7 years of dedicated cloud focus. Paul has helped some of the largest enterprise customers grow their business and improve their operational excellence. In his spare time Paul is a huge football and NFL fan.<\/p>\n

\"\"Mike Bezak is a Senior Technical Account Manager in AWS Enterprise Support. He has over 20 years of experience in information technology, primarily disaster recovery and systems administration. Mike\u2019s current focus is helping customers streamline and optimize their AWS Cloud journey. Outside of AWS, Mike enjoys spending time with family & friends.<\/p>\n","protected":false},"excerpt":{"rendered":"This post is cowritten by Laura Skylaki, Vaibhav Goswami, Ramdev Wudali and Sahar El Khoury from Thomson Reuters.…\n","protected":false},"author":2,"featured_media":128170,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[220,218,219,61,60,80],"class_list":{"0":"post-128169","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-ie","12":"tag-ireland","13":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/128169","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=128169"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/128169\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/128170"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=128169"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=128169"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=128169"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}