The New Jersey Devils have launched “Bott Stevens,” a custom AI chatbot designed to enhance digital fan engagement.
Named after Devils legend Scott Stevens, the chatbot is powered by Theta EdgeCloud’s (THETA) decentralized AI infrastructure and will be available on the team’s official website during the 2024-25 NHL season.
The AI agent will provide fans with real-time information on game schedules, tickets, stats, and merchandise. Utilizing Theta’s Retrieval Augmented Generation technology, Bott Stevens will pull data from official NHL sources, ensuring accuracy while avoiding misinformation from unverified sources.
You might also like: Crypto exchange Deribit warns against job scams using fake recruiter profiles
Fans can ask the chatbot a range of questions, such as upcoming match details or season statistics, and receive instant responses.
Bott Stevens will be powered by Theta EdgeCloud’s network of over 30,000 edge nodes and distributed GPUs, offering more than 80 PetaFLOPS of computation power.
This ensures scalability, even during peak demand, such as playoffs or major team announcements.
Beyond answering questions, the AI chatbot will offer historical highlights, game recaps, venue information, and updates on team events. Over time, it may expand to include predictive analytics for fantasy sports and interactive fan engagement tools.
You might also like: Privy secures over $40m in funding to expand crypto infrastructure
Q&A with Mitch Liu, CEO of Theta Labs
Q: How will Theta EdgeCloud’s decentralized design make sure the chatbot can handle peak demand during major events?
A: EdgeCloud’s decentralized design offers over 80 PetaFLOPS of on-demand distributed GPU compute power, combining both cloud and edge computing devices. Over 30,000 global edge devices enable us to handle peak usage over the baseline capacity provided by cloud-based servers, and critical in professional and esports industries where major events such as playoffs and finals draw the most fan interest.
Q: How will “Bott Stevens” pull accurate, real-time data from sources like the NHL API? What safeguards prevent outdated or incorrect information? How will you keep RAG from being pulled from crowd-sourced and potentially inaccurate sources?
A: The EdgeCloud back-end platform defines which data sources to pull data from, and this feeds into the real-time RAG database, so crowd-sourced or inaccurate sources can be avoided. As in most generative AI LLM models, the biggest challenge is not just getting accurate real-time data from trusted sources but how that data is cleaned, prepared and normalized so it can be efficiently and accurately incorporated into the model.
Q: How will you ensure data privacy and security for users interacting with the chatbot? Are you storing user data?
The chatbot does not store or maintain any user data beyond the user’s session, and unlike traditional chatbots like DeepSeek or ChatGPT, Theta chatbots are focused on industry specific content, for example only hockey related information for the NJ Devils.
Q: Are there plans to expand “Bott Stevens” with features like predictive analytics? Do you want to take the tech any further?
A: A big area of development for us now is extending the interactive RAG chatbot to become proactive agentic AI with various extendable actions, that’s independent and autonomous. For example, being able to interface with customer service and CRM systems to automatically create support tickets to improve user satisfaction, and AI agents that create “fantasy sports/esports teams” that can play against users to increase fan engagement and a step towards predictive analytics for talent identification, player valuation, and opponent analysis.
Q: What attracted Theta Labs to partner with the New Jersey Devils? What customization options are available for other teams or brands?
A: Theta EdgeCloud has gotten good traction with professional sports teams starting with the Las Vegas Knights and esports teams including FlyQuest and Evil Geniuses. The user interface is fully customizable and branded to be integrated into each team’s website, mobile apps, Discord and other social channels. More importantly, the real-time data source and APIs that feed into EdgeCloud’s RAG database is customized to domain-specific information, such as NHL hockey, esports games like Valorant, League of Legends, team/league specific information, and more.
You might also like: MultiBank Group rumored to be planning the launch of MBG token
Q&A with Marc Ciampa, Vice President of Content of the NJ Devils
Q: How will the team promote “Bott Stevens” to drive fan engagement and adoption?
A: To promote “Bott Stevens,” the Devils will integrate promotion across multiple platforms to encourage fan interaction, “Bott Stevens” capabilities, and raise general awareness.
Q: What types of curated content will the chatbot offer beyond stats and schedules?
A: Beyond providing statistics and schedules, “Bott Stevens” will offer curated content to enrich the fan experience, including historical highlights and event and venue information. “Bott Stevens” will answer questions about memorable moments from the team’s history, provide details on upcoming events, theme nights, etc. in the Devils and concert space, and offer comprehensive details about Prudential Center amenities. These amenities include accessibility services, bag policies, concession locations and ATM locations throughout the venue.
Q: How will you measure the chatbot’s success in enhancing fan engagement or sharing relevant and accurate stats/data?
A: We will measure success through user engagement metrics, accuracy rates, fan feedback and support load reduction. We plan to monitor the number of interactions, session durations, and repeat usage, as well as track the precision of information provided, ensuring that stats and data shared are correct and up-to-date. We will also collect user feedback through surveys to understand satisfaction levels and areas for improvement and find ways to measure the decrease in inquiries handled by human support, indicating the chatbot’s efficiency in addressing common questions.
Q: As AI fatigue is growing among customers, how will you make “Bott Stevens” different from other AI agents that your clients already ‘digitally’ meet on a daily basis?
A: Through continuous learning, human-like conversational tone, personalized interactions, and creating a chatbot ecosystem. We will implement machine learning algorithms that adapt to fan feedback and evolving interests, ensuring the chatbot remains relevant and engaging, as well as design the chatbot’s communication style to be friendly and relatable, enhancing the overall user experience. We look to leverage AI to tailor responses based on individual fan preferences and behaviors, making each interaction unique. Eventually, “Bott Stevens” will be available in our app and tied to other key aspects such as our concession guide, wayfinding and other things we might think of down the line.
