G2 Taps AWS Bedrock for AI Co-Pilot to Guide B2B Buying

In a significant AWS Marketplace G2 integration update for the B2B technology market, software marketplace G2 has expanded its long-standing partnership with Amazon Web Services (AWS) to launch an AI-powered buying assistant. This new feature, named “Monty,” leverages AWS Bedrock and generative AI to transform how businesses discover and select software. Instead of relying on keyword searches and manual filtering, G2’s 90 million annual users can now engage in a conversational dialogue, describing their specific needs in natural language to receive personalized, context-aware recommendations synthesized from over 2.4 million verified user reviews. This latest G2 AWS AI partnership news marks a deliberate shift from a static search model to a dynamic, consultative experience, directly addressing the information overload common in B2B software procurement.

Key Points

• G2’s new AI assistant, “Monty,” is built on AWS Bedrock, utilizing foundation models from providers like Anthropic to power its conversational capabilities.

• The system employs Retrieval-Augmented Generation (RAG) to ground its responses in G2’s proprietary database of 2.4 million verified user reviews, enhancing relevance and reducing model hallucination.

• This development positions the platform as a G2 software procurement co-pilot, designed to address the “analysis paralysis” that 72% of B2B buyers experience, according to Gartner.

• A key architectural component is Bedrock’s data privacy guarantee, which ensures that G2’s customer and review data is not used to train the underlying base models.

Architecting Intelligence: The Bedrock-RAG Foundation

At the core of G2’s new functionality is Amazon Bedrock, a fully managed service that provides access to a selection of foundation models (FMs) through a single API. This architecture allows G2 to experiment with and deploy models from leading firms like Anthropic, AI21 Labs, and Cohere without managing the complex underlying infrastructure. For its assistant, G2 is using models from the Claude family, which are recognized for their proficiency in dialogue and summarization.

The implementation of the G2 using AWS Bedrock and RAG technology follows a sophisticated multi-step process. When a user enters a natural language query, the system uses Retrieval-Augmented Generation to perform a semantic search across its vast review database. Rather than matching keywords, it finds reviews and product data that are contextually relevant. This curated information is then fed to the Claude model, which synthesizes it into a coherent summary and a ranked list of software recommendations, with justifications rooted directly in user feedback. This approach leverages Bedrock’s enterprise-grade privacy controls, which, as detailed by AWS, prevent proprietary G2 data from being used to train the original FMs—a critical requirement for maintaining data security and trust.

Unraveling the SaaS Maze

G2’s initiative directly addresses the increasing complexity of the B2B SaaS market, which is projected to exceed $880 billion by 2030 according to Fortune Business Insights. With the average organization now using 130 different SaaS applications, as reported by Productiv, buyers face a significant challenge. Gartner research highlights this “analysis paralysis,” noting that B2B buying groups involve six to ten decision-makers and that a majority of tech buyers now prefer a “rep-free” experience.

The “procurement co-pilot” model adopted by G2 fits into a broader industry trend of embedding AI assistants into complex professional workflows. This paradigm is already established with tools like GitHub Copilot for developers, Microsoft 365 Copilot for productivity tasks, and Salesforce’s Einstein Copilot for customer relationship management. By applying this model to software procurement, G2 is evolving its role from a simple aggregator of reviews to an intelligent synthesizer of information. The goal is to cut through the noise and provide the clear, data-driven guidance that modern buying teams require.

Grounding AI in Verified Reality

While the technical advancements are notable, the success of this generative AI for B2B software buying tool hinges on trust and transparency. As a platform built on verified reviews, G2’s primary asset is its credibility. The company addresses the inherent challenges of generative AI, including the potential for model “hallucination” and the perpetuation of biases found in training data. The use of RAG is a key technical choice to mitigate this, as it grounds the AI’s outputs in specific, verifiable review data rather than allowing the model to generate information freely.

According to G2 CEO Godard Abel, the objective is clear: “With 90 million annual buyers, we have a responsibility to make their software selection process not just easier, but better.” Success is measured not just by user engagement with the conversational interface but by tangible outcomes, such as reduced research time and improved satisfaction with software purchases. Maintaining transparency about how recommendations are generated remains paramount to ensuring users trust the AI’s output is objective and not a “black box” influenced by advertising.

Conversational Commerce: Beyond Keywords

The G2 and AWS partnership represents a concrete application of generative AI to a persistent business problem. It moves the B2B software discovery process beyond the limitations of the search bar and toward a more intuitive, conversational paradigm. This development demonstrates how AI is being deployed not to replace human decision-making but to augment it, providing a powerful tool to navigate an increasingly crowded and complex digital marketplace—a space where the AI in Marketing sector alone is projected to reach over $100 billion by 2028. This shift toward AI-driven hyper-personalization reflects a trend that McKinsey identifies as reinventing core business functions. As these AI co-pilots become standard, how will businesses adapt their procurement strategies to leverage—and critically verify—these new digital consultants?