Sage Taps AWS to Build Domain-Specific AI for Accounting

Sage has expanded its collaboration with Amazon Web Services (AWS) to develop AI solutions for accounting and compliance tasks. The partnership builds on Sage's existing status as an AWS partner with SMB Software Competency recognition.

The accounting software firm is utilising AWS AI services including Amazon Bedrock—a managed service that offers foundation models through an API—alongside purpose-built AI chips such as AWS Trainium and AWS Inferentia to accelerate development of financial management tools. 

This infrastructure combination enables Sage to test and deploy AI models specifically designed for accounting workflows.

Domain-specific models enhance trust in financial AI

Over the past year, Sage has developed Large Language Models (LLMs) and Small Language Models (SLMs)—neural networks trained on text data that can generate human-like responses—that focus on accounting and compliance use cases.

These specialised models power the Sage Copilot assistant, which is currently available in five markets including the UK, US, France, Spain and Germany. 

The assistant is embedded within several Sage products including Sage Accounting, Sage for Accountants, Sage 50, and Sage Intacct.

A key component of this technology is the Accounting Filtering Model, which ensures AI-generated responses remain within financial and compliance domains, preventing the system from providing inaccurate information. 

The company has also developed endpoint detection models that interpret user intent and retrieve financial data more precisely.

“Our collaboration with AWS enables us to develop and scale AI solutions that directly address the challenges SMBs face,” says Amaya Souarez, Executive Vice President of Cloud Services & Operations at Sage. 

“By combining Sage's deep accounting expertise with the powerful AI infrastructure of AWS, we are creating technology that is not just intelligent, but also trusted and practical for finance.”

Semantic search transforms financial data access

Beyond the accounting models, Sage has implemented a Semantic Search API powered by Amazon Bedrock that transforms how SMBs access financial information. 

The system enhances the retrieval and ranking of responses from financial and compliance data repositories.

The search capability reduces time spent navigating financial data and improves decision-making processes for businesses with limited resources. 

A self-service content hub allows Sage's product teams to refine knowledge sources, ensuring AI-generated responses reflect current financial regulations.

Sage is extending its semantic search capability to support multiple languages and serve its global customer base. This multilingual support forms part of the company's strategy to provide consistent services across different markets.

The system aims to help SMBs with limited resources reduce administrative burdens associated with financial management and regulatory compliance.

“These advancements make financial decision-making smarter, more intuitive, and more efficient, helping SMBs reduce time spent on financial admin, gain greater visibility into their finances, and focus on growing their business,” Amaya says.

As Sage continues to expand its AI capabilities, the company remains committed to responsible AI innovation, developing solutions that are transparent, secure, and designed for SMB needs.

This work lays foundations for advanced agentic capabilities within Sage Copilot, supporting more intelligent automation and proactive financial assistance.

Explore the latest edition of FinTech Magazine and be part of the conversation at our global conference series, FinTech LIVE. 

Discover all our upcoming events and secure your tickets today.

FinTech Magazine is a BizClik brand