The New Jersey Devils have launched Bot Stevens, a customized AI chatbot designed to boost digital fan engagement.
Named after Devils legend Scott Stevens, the chatbot options Theta Edgecloud (Theta) decentralized AI infrastructure and is on the market on the staff's official web site throughout the 2024-25 NHL season.
AI brokers present followers with real-time details about sport schedules, tickets, statistics and merchandise. Utilizing Theta's search and enhanced technology know-how, Bott Stevens extracts knowledge from official NHL sources to make sure accuracy whereas avoiding false data from unverified sources.
You would possibly prefer it too: Crypto Change Deribit makes use of pretend recruiter profiles to warn in opposition to employment fraud
Followers can ask the chatbot a wide range of questions, together with particulars of upcoming matches and season statistics, and get immediate solutions.
Bott Stevens options over 30,000 edge nodes and a community of distributed GPUs, offering over 80 PetaFlops computing energy.
This ensures scalability even in peak demand, similar to playoffs and main staff bulletins.
Along with answering questions, the AI chatbot offers historic highlights, sport abstract, venue data and staff occasions updates. Over time, we could develop to incorporate predictive analytics for fantasy sports activities and interactive fan engagement instruments.
You would possibly prefer it too: Privy has secured greater than $40 million in funding to develop its crypto infrastructure
Q&A with Mitch Liu, CEO of Theta Labs
Q: How does Theta Edgecloud's distributed design permit chatbots to deal with peak demand throughout main occasions?
A: EdgeCloud's distributed design combines each cloud and edge computing units to offer over 80 petaflop-on-demand distributed GPU computing energy. Over 30,000 international edge units can help you deal with peak utilization throughout the baseline capability supplied by cloud-based servers, and key occasions similar to playoffs and finals are necessary within the {industry} which are most attracting fan curiosity in skilled and esports industries.
Q: How does “Bot Stevens” extract correct, real-time knowledge from sources just like the NHL API? Which safeguards stop outdated or misinformation? How can I stop RAG from being pulled from crowdsourced, probably inaccurate sources?
A: EdgeCloud's backend platform defines the information supply from which it extracts knowledge. This feeds right into a real-time RAG database, so crowdsourced or inaccurate sources may be prevented. Like most generative AI LLM fashions, the most important problem shouldn’t be solely the way to get hold of correct real-time knowledge from dependable sources, however how that knowledge may be cleaned, ready, normalized and effectively and precisely integrated into the mannequin.
Q: How do you assure knowledge privateness and safety for customers interacting with the chatbot? Do you retailer consumer knowledge?
Chatbots don’t retailer or keep consumer knowledge past the consumer's periods, and in contrast to conventional chatbots similar to DeepSeek and ChatGpt, Theta Chatbot focuses on industry-specific content material. For instance, solely hockey associated data from NJ Devils.
Q: Are there any plans to develop “Bot Stevens” with options like predictive analytics? Do you need to get extra expert?
A: Presently, a significant space of growth is increasing interactive RAG chatbots to turn out to be proactive agent AI with a wide range of scalable actions. For instance, you possibly can interface with customer support and CRM techniques to routinely create help tickets to enhance consumer satisfaction. An AI agent additionally creates “fantasy sports activities/esports groups” that may compete in opposition to customers to extend the steps in the direction of fan engagement and expertise identification, participant analysis, and predictive evaluation of opponents.
Q: What impressed you to draw Theta Lab to associate with the New Jersey Devils? What customization choices can be found to different groups and types?
A: Theta Edgecloud has gained emergency with skilled sports activities groups beginning with the Las Vegas Knights and the esports staff, together with the FlyQuest and Evil Geniuses. The consumer interface is totally customizable and branded to combine into every staff's web site, cell app, Discord and different social channels. Extra importantly, the real-time knowledge sources and APIs that feed into EdgeCloud's RAG database are custom-made to domain-specific data similar to NHL Hockey, Valorant, League of Legends, and eSports video games, similar to Workforce/League-specific data.
You would possibly prefer it too: Multibank Group is rumoured to be planning to launch MBG tokens
Q&A with Mark Cianpa, Vice President of Content material at NJ Devils
Q: How will the staff promote “Bot Stevens” to advertise fan engagement and adoption?
A: To advertise “Bot Stevens,” the Devils will combine promotions throughout a number of platforms to advertise fan interplay, the flexibility of “Bot Stevens” and lift public consciousness.
Q: What sort of curated content material does a chatbot provide past statistics and schedules?
A: Along with offering statistics and schedules, Bott Stevens affords curated content material to complement the fan expertise, together with historic highlights and details about occasions and venues. “Bot Stevens” solutions questions on memorable moments from the staff's historical past, offers particulars on upcoming occasions on the Devils and live performance areas, theme nights, and extra, in addition to complete particulars on the Prudential Centre facilities. These amenities embrace accessibility providers, bag insurance policies, concession places, and ATM places all through the venue.
Q: How do you measure chatbot success in enhancing fan engagement and sharing related correct statistics/knowledge?
A: Measure success by consumer engagement metrics, accuracy, fan suggestions and decreased help load. It’ll monitor the variety of interactions, session period, and repetitive utilization, observe the accuracy of the knowledge supplied, and be sure that the shared statistics and knowledge are appropriate and updated. It additionally collects consumer suggestions by analysis to grasp satisfaction ranges and areas of enchancment, discover methods to measure the discount in inquiries dealt with by human help, and present the effectivity of chatbots in addressing frequent questions.
Q: With AI fatigue growing amongst clients, how do you make “Bot Stevens” completely different from different AI brokers that shoppers already meet “digitally”?
A: By means of steady studying, human-like dialog tones, customized interactions, and making a chatbot ecosystem. Implement machine studying algorithms that adapt to fan suggestions and evolving pursuits, making certain that chatbots proceed to be related and enticing, and design chatbots' communication kinds to be pleasant and related, enhancing the consumer expertise. Leverage AI to regulate responses based mostly on particular person fan preferences and conduct, making every interplay distinctive. Ultimately, “Bot Stevens” will probably be out there within the app, tied to different necessary points similar to concession guides, wayfinding and different concepts.
