New York Jets Image
New York Jets 1969 Super Bowl Win
In what ways can data and AI transform professional sports? Nearly a quarter century ago, the book Moneyball forever changed the perception of how data and analytics could be applied to gain competitive advantage in professional sports. Several years ago, I published an update here in Forbes on the current state of data and analytics in the National Football League (NFL) based on conversations with NFL Chief Data and Analytics Officer Paul Ballew.
Now, in the midst of the unprecedented and rapidly accelerating rate of AI adoption across industries and all sectors of society, I had the opportunity to speak to a long-time industry colleague and seasoned Chief Data and Analytics Officer (CDAO), Iwao Fusillo, who recently assumed the role as the first CDAO for the New York Jets. I still vividly recall how the New York Jets made history as the first team from the old American Football League (AFL) to win a Super Bowl–Super Bowl III in 1969—led by the legendary NY Jets quarterback Joe Namath (a.k.a. Broadway Joe).
Iwao Fusillo, a lifelong New York Jets fan, brings a long history of data, analytics, and more recently, AI leadership to his new role. For more than seventeen years, Fusillo held senior data and analytics leadership roles with American Express, including a tenure as global head of data strategy and insights. From there he went on to his first stint with the NFL, where he held the role of senior vice president for data & analytics. Fusillo subsequently served as chief data and analytics officer for General Motors and global head of data and analytics at PepsiCo. The range and depth of his data and analytics experience, and now AI, is matched by few.
The Data and AI Leadership Mandate for the NY Jets
Fusillo hopes to bring his experience in driving data, analytics, and AI leadership to bear for an iconic NFL franchise, with the early focus on building the data and AI foundation that supports better decisions across football and business. Fusillo notes, “The Jets have created the franchise's first Chief Data & Analytics Officer role, which I'm fortunate to hold and which, as far as we know, is the only such position across the 32 NFL clubs today.” Fusillo adds, “My mandate is to help build the best AI decision infrastructure in sports.”
“My job is to architect a single data-and-AI foundation that serves the whole organization and to act as a strategic partner to leadership across both football and business,” explains Fusillo. “When I joined the club, work was already underway. One hundred days later, AI is embedded in the daily activities of our front office – running in production across football and business operations and delivering measurable value to the business.” He comments, “Over time the biggest advantage of data and AI is compounding insight, where football decisions inform business strategy and business data sharpens our understanding of fan and market dynamics.”
Fusillo elaborates, “My remit is unusual in three ways: it spans both football analytics and business analytics — functions that are separated at every other team — it also includes application development (typically part of the CTO/CIO orgs in other companies), and of course AI strategy and deployment.”
Leadership for Data and AI Starts at the Top of the NY Jets Operations
Leadership at the NY Jets starts at the top of the organization. “Our owner, Woody Johnson, has given us a clear charge: make the New York Jets the best AI decision infrastructure in sports,” says Fusillo. He continues, “There was no playbook to follow. No NFL club had unified football analytics, business analytics, application development, and AI under one leader, and there was no peer model to study.”
“Woody provided our north star by analogy. He pointed to the way WWE and Broadway producers relentlessly focus on the entire live entertainment experience — the lighting, the sound, the choreography — while sports have historically surveyed fans after the game and adjusted for the next one,” notes Fusillo. “This new framing is now shaping how the whole organization thinks about game day.”
“The goal of becoming the best AI decision infrastructure in sports is a stated organizational priority set by our ownership — not a technology project living in a corner of IT,” explains Fusillo. “President Hymie Elhai, Chief Operating Officer Brian Friedman, and General Manager Darren Mougey each ensured a cross-functional mandate spanning both football and business, which is what has made rapid progress possible.”
Fusillo explains, “For decades, teams used data to explain what already happened — on the field, in ticket sales, in sponsorship performance.” He continues, “The shift we are driving is from analytics as hindsight to AI as decision-making infrastructure: embedding intelligence directly into the workflow so that coaches, scouts, sales leaders, and operators are supported by systems that surface trade-offs, risks, and second-order consequences before a decision is made — in near real time, and at scale.”
Fusillo concludes, “Our people are, and will remain, responsible for the decisions. Data and AI make them faster, more informed, and exponentially more scalable.”
Integrating Data and AI into Operations at the New York Jets
Today, the New York Jets have well over 20 AI applications that operate live across football and business operations, each with named owners and performance metrics. A few examples include:
Autonomous Agents -- On the business side, the Jets communications team runs a fully autonomous agent that publishes a comprehensive Jets media-intelligence report every day, giving team executives early signals to get ahead of stories rather than learning about them in a meeting three days later.
Mining CRM and Public Data – The Jets sponsorship and premium-sales teams use AI to mine CRM and public data to surface prospects that are one or two degrees of separation from the team, to draft first-pass pitches.
Using Anthropic’s Claude – The Jets are using Claude for deeper research studies — ranging from CRM customization to ticketing strategy. Ahead of this year’s schedule release, the team used Claude to run over 100,000 Monte Carlo simulations to stress-test variable-pricing tiers, letting the team lock in pricing in hours instead of days.
AI Agents in Football Operations -- Ahead of the recent NFL Combine, the team stood up an AI agent, built with Google Gemini, that extracted critical information from the hundreds of pages of medical dictation generated for roughly 340 prospects. Work that used to take weeks of manual effort now happens in about a day, with more consistent output.
AI in Coaching and Scouting – New features in the Jets’ in-house, cloud-native application for coaches and scouts, called Titan, are launching in days instead of weeks through the use of software-engineering coding assistance. In one case, a coach photographed a user interface he liked in a different context, and our team built a working prototype in 30 minutes.
Fusillo explains, “Because we are a Microsoft shop, Microsoft Copilot was the natural place to start. We went from a standing start to 91% Copilot adoption across our entire non-player front office in 100 days, averaging 62 prompts per active user per month.” He continues, “Our data engineering and data science teams recently built a pipeline of three years of Delaware North concessions data — 2.4 million transactions, millions of dollars in value, 5 million items — into mobile-ready insights for our commerce team in 24 hours.”
“Combined with our new partner CrowdIQ's computer-vision feeds measuring crowd flow and fan sentiment section by section, and normalized against the play-by-play, we are building a real-time command center for the in-stadium experience — and a fundamentally new, verifiable basis for serving fans and partners alike,” notes Fusillo.
The bigger shift for the Jets is moving from what the industry calls bolt-on AI — where an agent is simply a faster pair of hands on an existing workflow — to AI-native workflows that are rebuilt around agents. Fusillo notes, “This is how a 2-to-3x multiplier becomes a 20-to-30x one.” He continues, “We are building game day into a real-time data product — one that helps us serve fans better in the moment and gives our partners a more verified, transparent picture of engagement — and moving sponsorship from retrospective, survey-based valuation to verified, in-the-moment activation.”
Building a Data and AI Organizational Culture that Delivers Results
Business value from AI is measured by the Jets through a three-horizon framework that the organization uses to operationalize AI. Fusillo comments, “Culture is Horizon 1, and it is the step most organizations are tempted to skip. Skipping it is a mistake for two reasons: you never bring your people to a common baseline of AI proficiency, and you lose your single best source of high-value ideas — the people who actually do the work are the ones best equipped to identify where AI creates double-digit returns.”
Fusillo notes that each horizon delivers a different kind and magnitude of return. The components are:
Embedding AI into daily work — This means putting Microsoft Copilot into the hands of the front office. The gains here are real, typically single-digit percentage improvements in productivity.
Delivering true workflow automation – This is where the organization targets and produces double-digit gains in revenue and productivity.
Establishing AI as decision architecture — This means establishing hybrid teams of people and AI agents — where the organization realizes multiplicative returns on the order of 2 to 3 times today with more to come.
“The clearest way to see the value is in compressed time and cost,” says Fusillo. “We have let the business lead rather than mandating adoption from IT, which has built genuine trust. We have invested heavily in education — working with our partner Next League on workshops, office hours, and a clear curriculum covering what AI is, what it isn't, and its real risks.” He adds, “The unified football-and-business remit is itself a value multiplier: one data foundation, one AI platform, and one set of guardrails serving both sides effectively doubles the return on every shared infrastructure investment.”
“A football data-foundation effort that took roughly eight months and nearly 100 contractors in a comparable build I led earlier in my career at General Motors, is something that our team at the Jets completed in eight weeks using Google Gemini.” He adds, “Medical review ahead of the NFL Combine, which used to consume our staff for three to four weeks, now takes about a day. A coaching-and-scouting dashboard change that used to take weeks or months, our developers rebuilt in roughly 30 minutes.” Fusillo continues, “Our software and data science teams produced the equivalent of about four human-years of coding in 90 days using Anthropic's Claude Code and have stood up our modeling data pipeline in weeks rather than months with OpenAI's Codex.”
Fusillo concludes, “Our leadership showed up in person — the informal time we spent alongside staff across the building, including regular lunches across departments, mattered as much as any formal program. The proof is in the engagement: not only 91% adoption, but average usage that more than doubled to 62 prompts per active user per month, and 65 business use cases surfaced by our own staff. People aren't just enrolled — they're relying on it.”
Ensuring Responsible Use of AI Across the NY Jets Organization
Responsible AI is built into the Jets AI execution plans at every stage. Fusillo comments, “We are building AI governance in parallel with deployment. We are focused on who certifies our AI agents, how we track which models power which workflows, and how we match the right people to the right level of AI work.”
Fusillo cites the example of the Jets four-layer prompt-evaluation discipline, which comprises:
Context Layer – This ensures the model knows what sources to draw on and what role to play, which is essential to preventing hallucinations.
Bias Layer – This keeps prompts specific enough to be actionable but not so specific that the prompt itself introduces bias.
Constraint Layer – This defines the required output structure and explicit exclusions.
Evaluation Layer – This instructs the model to check itself, cite its sources, and report confidence levels back to the user.
Reflecting on the human impact of AI, Fusillo notes, “One question that comes up constantly is whether AI is killing entry-level jobs. Our perspective is hopeful, but also more demanding.”
Fusillo continues, “That's also why we are optimistic, not fearful, about early-career talent. AI doesn't shrink what a young analyst can contribute — it expands it. The entry-level hire who arrives fluent in these tools can take on work that used to require years of experience, which makes them more valuable, not less.” He adds, “The people who become AI-native will be the ones who get hired; those who don't risk being left behind.”
“We keep returning to a question from Stripe’s co-founder and CEO Patrick Collison: what would we choose to do much more of, if we suddenly had the leverage?” notes Fusillo. “For us, AI isn't about where we can cut jobs — it's about executing the opportunities we never had the capacity to pursue. That is the version of the Jets we are building, together.” He adds, “Underlying all of this is a principle we keep coming back to: we are not automating judgment or decision-making — we are automating the work around decisions. AI does the research, the synthesis, and first-pass drafting; our people decide who to call, what to offer, and how to close.”
“The organizations that get the most from AI won't be the ones with the flashiest models. They'll be the ones that build trusted, governed systems that help their people make better decisions every day — faster, with less bias, and with clearer accountability,” concludes Fusillo.