Opening & Welcome
Andrew Delaney, President & Chief Content Officer, A-Team Group 

Practitioner Keynote: From pilot to payoff: Navigating the journey to scalable, trustworthy AI

• From unstructured to usable: How can GenAI turn chaotic text into crisp signals, unlocking business rich insights that have historically been invisible to the enterprise?
• Trust by design: How can teams navigate the shift from pilot to production—moving fast without breaking trust, keeping governance non negotiable while letting innovation breathe?
• Defining ‘Good’ ROI: How can firms move beyond soft metrics to create a repeatable framework that captures both efficiency gains and the harder to quantify value of risk reduction?
• From skeptics to sponsors: How do you ignite business excitement that turns leaders into co?owners, so adoption is pulled, not pushed, and the path from pilot to production is genuinely business led?

Fireside Chat with:
Jennifer Ippoliti, Legal Chief Data Officer, JP Morgan Chase
Interviewed by: Jane Conway, MD, Digital, Data and enablement – Client & Product Solutions, Apollo Global Management

User C Level Panel: The AI value mandate – A CDO Playbook for measuring and delivering ROI 

• Measuring what matters: What are the most effective KPIs for measuring AI’s business value at each adoption stage: from initial experimentation to enterprise-wide transformation?
• Organising for value: Beyond simple collaboration, what new operating models or ‘fusion teams’ are proving most effective for aligning data, tech, and business units to rapidly prototype and scale high-value AI solutions?
• Finding the evangelist: How do you identify and empower the right business-line talent who can act as a driving force and an AI evangelist to champion initiatives from the ground up?
• Justifying the ‘probabilistic’ bet: How do you build a defensible business case for AI, and what’s a tangible example of quantifying a probabilistic benefit (like improved insight) versus a deterministic cost saving?
• Calculating the ROI of governance: As a data leader, how do you balance the demand for rapid AI innovation with the non-negotiable costs of governance and control? How do you articulate the ROI of not having a model fail in production?
• Managing the value pipeline: How do you strategically prioritise your firm’s portfolio of AI projects, and what is your framework for ‘failing fast’ and decommissioning initiatives that don’t deliver their promised business value?

Moderator: Julia Bardmesser, Adjunct Professor; NYU Stern School of Business; Founder and CEO, Data4Real
Andrew Foster, Chief Data Officer, M&T Bank
Sherry Marcus, Head of AI, TradeWeb
Michael Kaufman, US Chief Data Officer, Scotiabank 

Keynote: Informatica

A-Team Research Report Update: AI adoption in financial services – strategic implications for the office of the Chief Data Officer
Andrew Delaney, President and Chief Content Officer, A-Team Group

Panel: Holistic AI Governance – From black box to business value

• As governance expands beyond model validation, how must the partnership between the CDO, CISO, and Chief Risk Officer evolve to build a single framework that covers Model Risk, bias, and data security?
• How can firms protect sensitive data against specific attacks like prompt injection and who ultimately owns the risk of leakage in GenAI applications, the data team or cybersecurity?
• How can firms use AI as an active tool to automatically scan governance metrics and KPIs, pinpointing data quality hotspots and focusing resources where they are needed most?
• What is a practical, human led test for detecting bias in opaque vendor models, and what new forensic skillsets do governance teams need to perform this analysis?
• How can firms balance the need for powerful black box models with the regulatory demand for explainability, especially when an AI decision impacts customers financially?
• When an AI system fails—either by leaking data or making a biased decision—where does the ultimate accountability lie, and what investments are needed to ensure teams can effectively challenge these models?

Moderator: Marla Dans, Chief Data Office, Head of Data Governance, Formerly Chicago Trading
Cheryl Benoit, Executive Director, Operational Risk, Mizuho
Arun Maheshwari, Head of Model Risk Control, Legal and Compliance, Morgan Stanley
Peggy Tsai, AI and Data Product Director, JP Morgan

Keynote: Unlocking the keys to the kingdom – managing and governing unstructured data 

Panel: Architecting the intelligent ecosystem: AI as the blueprint and the builder

• How can AI help firms break down data silos and integrate data and where are firms on their journey?
• Unlocking value from unstructured data opens the keys to the kingdom. Beyond storage, what specific AI/ML models are most effective for automatically classifying, tagging, and extracting structured insights from unstructured data at scale?
• How can GenAI and RAG (Retrieval-Augmented Generation) be used to automatically map, model, and generate the enterprise semantic layer, drastically reducing the manual effort required?
• Given that mainframes and legacy systems aren’t disappearing, how can AI be used to create intelligent abstraction layers or “digital twins” of these systems, making their data accessible without costly and high-risk rewrites?
• What is the strategic architectural framework for the ‘build vs. buy’ decision? When should a firm build its own custom AI models vs. integrating a vendor’s AI CoPilot or specialized platform?
  
• Looking forward, what is the biggest architectural shift required to support Agentic AI? How do we design an ecosystem where AI agents are not just users of data but are trusted to actively and autonomously manage the data landscape itself?

Moderator: Brian Buzzelli, Director, Data & Digital Transformation, Meradia
Steven O’Bott, Chief Data Architect, Vanguard
Akshay Pore, MD, Data Modernization, AI Automation & Strategic Architecture, Bank of America

Real world AI use case: Deploying AI agents at Northern Trust to automate reconciliation and empower data users

• How can AI agents be deployed to automate complex reconciliations and empower internal users to find and analyze the data they need faster?
• How should firms build a robust control framework and guardrails to manage the non-deterministic nature of AI and ensure data accuracy?
• What are the tangible benefits of such an approach such as reducing data reconciliation times and automating manual data processes?

Suemee Shin, SVP Core Data Services Product Management, Northern Trust Asset Management

Panel: The intelligent data marketplace: From static products to AI-powered agents

• How are conversational AI and agentic interfaces solving the ‘last mile’ problem between data producers and consumers, allowing business users to ‘talk’ to data products directly to get answers, not just raw data?
• Beyond just creating tags, how can GenAI be used to autonomously package unstructured data (e.g., news, research filings) into monetizable, themed data products with summaries, key entities, and sentiment analysis already included?
• Achieving frictionless governance: How can firms move to a frictionless governance model using AI to automate the risk of a data request, to grant real-time, policy-driven access to intelligent data products?
• For the vendors: How are you embedding AI and LLMs into your marketplace platforms to provide not just a catalog, but intelligent data discovery, usage recommendations, and proactive quality alerts for listed products?
• When a data product is an interactive ‘agent’ rather than a static dataset, how does that change the way we measure ROI? Do we move from tracking ‘downloads’ to tracking the value of the decisions enabled by the agent’s insights?
• What is the biggest cultural shift for an organization to start thinking of its data teams not as report builders, but as AI product managers who develop and maintain a fleet of intelligent data agents for the enterprise?

Moderator: Dessa Glasser, Independent Board Member, Oppenheimer & Co.,
Linda Zhang, Executive Director, Commercial and Investment Banking, Data & Analytics Technology, JP Morgan Chase & Co
Fabien Thiaucourt, SVP, Data Governance & Enablement, Mastercard
Matt Katz, Field CTO, Arcesium

Panel: From rulebook to report – Applying GenAI to automate regulatory compliance

• How can Generative AI be used to interpret new regulations and rule changes, reducing the manual effort for compliance?
• How are firms using AI to automate data lineage required for regulatory reports and to perform faster root-cause analysis when a quality issue is flagged?
• What are the tangible use cases for AI in the reporting lifecycle and how have you measured success? 
• What does the optimal “human-in-the-loop” workflow look like when using AI to generate a regulatory filing? Where are the critical checkpoints for compliance and legal teams to review and approve AI-generated content to ensure accuracy and accountability?
• For the vendors: How are you ensuring your AI models are kept up-to-date with the latest regulatory interpretations and guidance, and how do you provide explainability to a client and their auditor for an AI-generated number?
• Looking forward, could reliance on an AI to generate regulatory reports be considered an outsourcing or model risk? What new due diligence and governance frameworks are needed to manage these AI-powered RegTech solutions?

Moderator: Chris Doris, Executive Director, Morgan Stanley
Brian Greenberg, Senior Director – Business Engagement Lead for Enterprise Data Management, BNY
Peter Gargone, Founder & CEO, n-Tier Financial Services

Champagne Roundtables

Join a roundtable discussion for a deep-dive, interactive discussion with your peers on some of the day’s most important themes. Address common problems, benchmark your progress and come away with practical solutions and takeaways!

Roundtable 1: How to ensure high quality and trusted data for AI

• What does an enterprise-grade data quality framework look like in an AI environment where small errors can compound into large model failures?
• How can firms automate data quality monitoring, enrichment, and reconciliation to support real-time AI workloads?
• What processes and controls are required to build trust in AI training datasets, metadata, lineage and validation?
• How do you measure and communicate the ROI of improved data quality, especially when value is realised through reduced risk rather than direct cost savings?

Host: Ellen Gentile, Former Enterprise Data Quality Team Leader, Edward Jones

Roundtable 2: Managing and governing unstructured data

• What are the most effective discovery methods for identifying, classifying, and securing unstructured data across multiple repositories?
• How should firms align unstructured data governance with compliance and communications monitoring requirements, especially under increasing scrutiny from regulators?
• What does AI readiness look like for unstructured data such as cleaning, labelling, normalising, and creating training sets?
• How can firms implement robust controls, observability, and auditing across AI pipelines that use unstructured data, particularly for bias, drift, and prompt-related risks?

Host: Julia Bardmesser, Adjunct Professor; NYU Stern School of Business; Founder and CEO, Data4Real

Roundtable 3: Operationalizing data products at scale

• How do firms define repeatable standards for designing, governing, and measuring data products across business lines?
• What architectural enablers (data platforms, pipelines, metadata, governance layers) matter most when scaling from dozens to hundreds of data products?
• How can catalogs, marketplaces, and self-service tools accelerate adoption while maintaining trust, compliance, and quality?
• How should firms incorporate external or alternative datasets into enterprise data products without compromising control, lineage, or semantic consistency?

Host: Gurprit Singh, Former Global Head of Data & AI, Partners Capital 

Roundtable 4: Workforce education and training for the AI era

• Which roles and skillsets must evolve first as AI moves from experimentation to embedded enterprise capability?
• What training frameworks actually work: hands-on labs, scenario-based learning, AI literacy programs, or model governance certifications?
• How should firms balance centralized AI expertise with decentralized citizen AI capabilities across business teams?
• How do organizations build a cultural foundation that encourages responsible experimentation while maintaining compliance and control?

Host: Dessa Glasser, Independent Board Member, Oppenheimer & Co

Roundtable 5: Building the next generation data architecture and intelligent data ecosystem

• What architectural patterns support AI-first operations, data mesh, knowledge graphs, semantic layers, intelligent pipelines?
• How can firms modernize data warehousing and platform infrastructure to handle AI-driven workloads without disrupting existing systems?
• What role will knowledge graphs and semantic technologies play in enabling explainable, interoperable, and machine-navigable data?
• How should organizations evaluate and integrate intelligent data services, such as quality automation, observability, and AI-driven lineage?

Host: Brian Greenberg, Senior Director – Business Engagement Lead for Enterprise Data Management, BNY

Roundtable 6: AI sourcing dilemmas: Buy vs. build

• What frameworks help firms decide when to build custom AI models versus adopting vendor-embedded AI features or co-pilots?
• How do you compare total cost of ownership from data preparation to skill requirements across build vs buy scenarios?
• What are the key risks of vendor dependency, model opacity, and governance challenges when buying AI?
• How can firms structure hybrid approaches (build the core, buy the accelerators) to maximize innovation while controlling risk?

Host: Reinaldo Ynchaustegui, Former Chief Technology Officer, Asset Management Firm

Networking drinks reception