Responsible Autonomous AI. Safely Enforced.
Friday, October 23, 2026 | 09:00am Central | Austin, TX
The executive panel at 09:50am Central streams publicly on LinkedIn. No invitation needed for that session.
The year AI stopped recommending and started acting
AI systems are no longer just producing outputs for a human to approve. They are initiating payments, routing claims and executing decisions inside regulated workflows. The control model most enterprises built for predictive models was not designed for that.
RAISE 2026 is the Responsible AI Institute’s annual working session for its member community. Enterprise leaders, practitioners, researchers and partners come together to compare what they have learned, pressure-test shared approaches and decide what the Institute should build next.
No single institution should have to solve every risk, control, assurance and implementation problem on its own. This is a do tank, not a think tank. Members do not attend presentations. They contribute experience, test ideas and set priorities for the year ahead.
Learn from peers
Shape what comes next
Vote on which shared-infrastructure problem the Institute takes on first, and which offerings get prioritized in 2027.
Strengthen your own program
Leave with practical guidance, assessment approaches and benchmarks your internal teams can use immediately.
Build credible leadership
Be visible as an active participant in an independent nonprofit advancing rigorous AI governance across industries.
180 minutes, end to end
The program runs in two halves. The first is open, and includes an executive panel streamed publicly on LinkedIn. The second is closed to Institute members and invited guests, so the room can work without an audience.
To registered attendees, in person and virtual.
Open to anyone. No invitation needed.
The second half of the program. Not streamed.
Opening, Welcome and the State of Responsible AI
The Institute frames the shift defining the day: AI is moving from generating outputs to making decisions and taking actions, and enterprises need a control and assurance model that matches. One open question goes to the room, and the closing session answers it.
A shared understanding of why autonomous AI requires a different assurance model.
What RAI Institute Members Built and Advanced in 2026
A state of the Institute address covering membership growth, expanding research partnerships including the incoming U.S. Bank funded fellows at Imperial College, Cambridge and Stanford, and progress across the TrustX Healthcare and Finance working groups. Built around the year's two or three defining themes, with a real member example inside each one.
What changed in the last twelve months, and what members can draw on now.
Live TrustX Member Sandbox: From Agent Idea to Production Readiness
One scenario, an autonomous claims-approval agent, carried end to end through all six steps of the TrustX framework: Identify, Classify, Control, Test, Verify and Improve. A working walkthrough rather than a framework overview, so the panel that follows lands against something concrete.
A firsthand view of what it actually takes to move an agent from idea to production readiness.
Who Controls Autonomous Finance and Healthcare?
The flagship discussion. Leaders from banking, asset management and healthcare examine what changes when AI systems move from recommending to acting: decision authority, human oversight, escalation paths and the evidence boards and regulators will expect before an autonomous system reaches production. Questions from the LinkedIn stream come into the room for the closing Q&A.
- What prevents promising autonomous AI pilots from reaching production?
- Which controls become essential once AI can act?
- Where must human approval stay mandatory?
- How should institutions handle failures, drift and unexpected behavior?
- What proof should executives, boards and regulators expect?
Member Connections and Public Stream Close
Informal exchange among members and speakers, and a clean close to the public broadcast before the member session begins.
Member Working Session: Shaping RAI Institute Offerings and the 2027 Roadmap
A facilitated session sized for the room. Leads from U.S. Bank, NatWest, Wellington Management and the Institute open each prompt, then the floor opens to everyone. The session maps where individual institutions are duplicating effort against where shared methodologies, sandbox environments and evidence standards could move the industry forward together. It closes with a live prioritization vote.
- Where does individual enterprise work create unnecessary duplication?
- What can institutions learn safely through shared sandbox environments?
- How should trust be communicated across customers, suppliers and regulators?
- What common evidence is needed for autonomous AI?
- How can nonprofit-led working groups accelerate responsible deployment?
A direct vote on which shared-infrastructure problem the Institute takes on first, and which offerings it prioritizes in 2027.
New Frontier: Enterprise Sovereign AI Initiative
The Institute introduces TrustX, Sovereign AI as its second horizontal module, sitting alongside the Finance and Healthcare working groups. The session frames sovereignty across data, model, infrastructure and operations as the next shared-infrastructure problem for members to help shape.
An early look at the Institute's next horizontal module, and how to get involved at the start.
The Next 12 Months: From Member Priorities to Action
The roadmap for the year ahead, opening with a direct response to the vote taken twenty minutes earlier. Covers new working group formation, sandbox development and how members can get involved early.
A concrete picture of what is coming and where to plug in.
Closing Call to Action
Back to the question asked in the opening, a synthesis of the day's conclusions and a direct ask: join the Institute, join a working group, stay engaged as the roadmap unfolds.
See Responsible AI Applied, Not Just Discussed
The TrustX sandbox is where members examine how an AI system behaves, what authority it holds, what controls its risk level requires and what evidence exists before it reaches production. At RAISE it runs live, against one autonomous agent, start to finish. TrustX produces independent verification, drawing on the 1,700 plus AI governance controls the Institute has maintained since 2016.
Define what the system does and what it is intended to do
Assess autonomy, authority, persistence, reach and impact
Map risk-proportionate controls to how the system actually operates
Exercise the system against defined risks and failure conditions
Produce evidence for internal review and external scrutiny
Refine controls as the system and its environment change
The Leaders Responsible for Turning AI Principles Into Practice
The program includes perspectives from AI leaders at major financial institutions, asset managers and the Responsible AI Institute.
U.S. Bank
NatWest
Wellington Management
Responsible AI Institute
Final speakers and titles will be announced before the event.
Avivah Litan is a recognized authority on artificial intelligence, trust, risk, security, identity and emerging technology. At RAISE 2026 she moderates the public executive panel and facilitates the member working session.
Built for the People Who Are Accountable When AI Acts
If your name is on the governance of a system that can make a decision without a human in the loop, this session is for you.
Â
- Chief AI Officers and enterprise AI leaders
- Chief Risk, Compliance and Legal officers
- Model risk and validation leads
- Internal audit and assurance leaders
- Technology and platform executives
- AI engineers and architects
- RAI Institute members and working group participants
- Prospective members, partners and implementation teams
The Leaders Responsible for Turning AI Principles Into Practice
The program includes perspectives from AI leaders at major financial institutions, asset managers and the Responsible AI Institute.
Questions about attendance or accessibility: hello@responsible.ai
Two ways to take part
In-person attendance is by invitation and the room is small. The executive panel is open to everyone.
Request an in-person seat
Watch the executive panel
Who Controls Autonomous Finance and Healthcare? streams live on LinkedIn at 10:50am Central. No invitation required. Bring questions, a team member is feeding them into the room.
Frequently asked questions
Who can attend RAISE 2026?
In-person attendance in Austin is by invitation, for Responsible AI Institute members and invited guests. The 45-minute executive panel streams publicly on LinkedIn and is open to anyone.
Is RAISE 2026 in person or virtual?
Both, with different access. The full three-hour program runs in person in Austin. The executive panel is broadcast publicly. The closed member working session in the second half is not streamed.
Will sessions be recorded?
[RECORDING POLICY]
Is this only for financial services?
No. Finance and healthcare are the Institute’s two active TrustX tracks, and Sovereign AI is being introduced at RAISE. The questions on the table, decision authority, oversight, evidence and audit readiness, apply across every regulated sector.
I am not a member. Should I still request a seat?
Yes. Prospective members and partners attend, and a portion of the room is reserved for organizations evaluating membership. Note your status on the form and the Institute will follow up.
The Next Era of AI Will Be Defined by What We Allow It To Do
Join the leaders building the boundaries, controls and evidence standards that make autonomous AI safe to deploy