TL;DR

The Royal Navy has deployed Atlas, an AI-powered avatar using large language models and RAG technology, to handle initial recruitment queries for submarine service. The system achieved a 76% reduction in human workload whilst maintaining a 93% satisfaction rate across 460,000 queries, demonstrating how public sector organisations can scale recruitment operations without compromising candidate experience.

AI Avatar Handles First-Line Recruitment Queries

The Royal Navy has launched Atlas, a real-time AI avatar powered by large language models, to manage initial contact with prospective submariners. The deployment represents a strategic evolution from text-based automation to immersive visual engagement, specifically targeting younger demographics who interact differently with digital channels.

Atlas builds on proven foundations. The Navy’s previous text-based assistant, recently upgraded to a full LLM with retrieval-augmented generation (RAG), processed over 460,000 queries from more than 165,000 users. That system achieved a 93% satisfaction rate whilst reducing live-agent workload by 76% and generating 89,000 expressions of interest—validating the approach before scaling to visual interaction.

The multi-vendor architecture, led by WPP Media’s Wavemaker for conversational design, delivers more than policy recitation. Atlas provides spoken answers, on-screen captions, and relevant videos from serving personnel when candidates ask about submarine life—a notorious conversion pain point. The system integrates directly with the NavyReady app and Enterprise CRM programme, ensuring data continuity across the recruitment funnel.

Paul Colley, Head of Marketing at the Royal Navy, emphasises the augmentation focus: “When it comes to AI, our focus is on how we can use it responsibly and strategically to better arm the teams we have. It’s not about replacing human support. It’s about giving the best support we can wherever and whenever candidates need it.”

Looking Forward

The Atlas deployment demonstrates a mature generative AI adoption pattern: start with data and simpler interfaces, validate efficiency gains, then scale to resource-intensive implementations. This test-and-learn approach positions the Navy to understand how conversational AI transforms candidate engagement whilst filtering low-value queries at scale, allowing human recruiters to focus on serious candidates. The trial’s results will inform broader public sector AI recruitment strategies.

Source: AI News