Karsten Wenzlaff, Advisor
August 26th, 2025
Karsten Wenzlaff is a Crowdfunding and Fintech Expert from Germany. He is Secretary-General of Digit[...]
C:\>DIR Hackathon 2026 Winners Put Agentic AI to Work
September 30, 2026 | NCFA Market Activity | Artificial Intelligence And Data, Risk Compliance And Regtech, Cybersecurity And Fraud, Digital Identity And Trust, Payments And Money Movement, Digital Assets Blockchain And Tokenization
Agentic AI Hackathon Winners, Financial Supervision And The Next RegTech Layer
On September 30, 2026, the C:\>DIR Global ‘Agentic Regulator’ Hackathon announced its 2026 winners and finalist demos after attracting 942 individual builders and 336 team submissions from more than 65 countries. Thirty six teams reached the final prototype round, including 19 from or affiliated with regulatory authorities, with projects covering market manipulation, fraud, agentic payments, smart contracts, AI advice and digital identity.
NCFA supported the inaugural hackathon as an Ecosystem Partner. What makes the results notable is that both regulators and builders spent a week turning emerging questions around autonomous financial systems into working prototypes.
Every finalist submitted code or a working application for assessment. 260 verified regulators and policymakers from more than 80 institutions across 40+ countries then selected the Final 6, which presented their solutions to 18 senior global regulatory leaders at the C:\>DIR Summit in Cambridge.
The primary question propelling the competition was how do regulators supervise software that can act?
Market Risk, Fraud and Agentic Payments Winners
1. 🥇 Team Pheonix, Securities and Exchange Board of India (SEBI), won first place for its work on market manipulation, agentic herding and financial stability. Its digital twins and trading agents are designed to let authorities "pressure test" proposed regulations in a synthetic securities market before they take effect.
2. 🥈 YU-SAM POWER, University College London, took second place with an agentic framework for assessing smart contract risk and identifying attacks, supporting more consistent enforcement and international collaboration.
3. 🥉 Rwanda Anti-Scam Team, with participants from the National Bank of Rwanda, Bank of Namibia, Proto, FNA, GLEIF and the University of Manchester, placed third with a system that uses agents to report and recover funds lost through fraud and scams in instant payment environments.
4. Team Axiom, from the State Bank of Pakistan, placed fourth with a supervisory intelligence system for agentic payments designed to identify systemic risks including agentic herding.
5. Agent Reliability Profile, from the MLCommons Financial Services Working Group, placed fifth with a standardized, testable reliability label for AI agents. Its central question is straightforward. Can an AI agent be trusted?
6. Probe, from the UK, placed sixth with a regulatory tool that uses an “Adversary” agent posing as realistic customers and a “Judge” agent to run thousands of test conversations against AI advice systems and flag potential regulatory breaches.
The vouch.finance Award went to Legis Labs, whose Agent Passport gives banks a way to check an AI agent's authority before a payment executes. Project Zeus, from the Dominican Republic's securities regulator SIMV, received an Honorary Award for a human governed, multi agent approach to detecting investment scams, engaging suspected scammers and gathering evidence for enforcement.
The six finalists shared recognition from a US$100,000 total prize pool. The Final 6 were also given an opportunity to demonstrate their work at the Singapore FinTech Festival.
Agent Identity Is Becoming a Financial Control
The projects tackled different problems, but several requirements kept appearing.
If an AI agent sends money, a financial institution needs to know which agent acted, who authorized it and whether the action stayed within that authority. If multiple trading agents begin behaving alike, supervisors need visibility into whether that behaviour is creating market risk. If an AI adviser gives unsuitable guidance, there needs to be an auditable record of what happened and where human oversight applied.
That helps explain why agent identity, traceability, auditability, human oversight and pre execution controls feature across so many of the finalist projects.
These aren't abstract governance principles once agents start touching payments, markets or regulated advice. They become operating requirements.
The six challenge areas made the same point from different directions:
- AI driven fraud and scams
- Agentic payments and commerce
- Decentralized market infrastructure and smart contracts
- Know Your Agent and digital verification
- Market manipulation and agentic herding
- AI enabled financial and high stakes advice
A New RegTech Layer Is Starting to Emerge
The commercial question is what happens after the prototype.
Putting these systems into production will require real data, cybersecurity controls, privacy safeguards, legal authority, model governance, procurement and integration with existing supervisory infrastructure. Cross border applications add another layer because an agent may operate across multiple firms, payment systems and jurisdictions.
That's where the opportunity becomes more interesting.
RegTech has traditionally focused on helping financial institutions interpret rules, monitor activity and demonstrate compliance. Agentic supervision creates another market for technology that helps regulators simulate markets, verify AI agents, monitor autonomous activity, test financial advice and investigate fraud.
Canada already has active policy and infrastructure work across AI, payments, digital identity, fraud and financial market oversight. Canadian fintech and RegTech companies working in these areas should be watching which of these prototypes move into pilots, standards or production environments.
NCFA covered the launch of the C:\>DIR Agentic Regulator Hackathon in July. The winners now provide an early view of what builders and regulators chose to build when given a defined supervisory problem, sandbox environments and a week to produce working solutions.
Strategic Takeaway
The interesting result isn't simply that regulators used AI at a hackathon. It's what they built.
They built systems to test markets, identify fraud, verify agent authority, monitor payments, assess smart contract risk and scrutinize AI generated advice.
As financial systems add more autonomous actors, the supervisory layer will have to become more capable too.
Agentic finance may require agentic supervision. The C:\>DIR Hackathon has now put some early working examples on the table.
Explore: C:\>DIR Hackathon 2026 winners, presentations and demos
Read: Official C:\>DIR Global Agentic Regulator Hackathon winners announcement
The National Crowdfunding & Fintech Association (NCFA Canada) is a financial innovation ecosystem that provides education, market intelligence, industry stewardship, networking and funding opportunities and services to thousands of community members and works closely with industry, government, partners and affiliates to create a vibrant and innovative fintech and funding industry in Canada. Decentralized and distributed, NCFA is engaged with global stakeholders and helps incubate projects and investment in fintech, alternative finance, crowdfunding, peer-to-peer finance, payments, digital assets and tokens, artificial intelligence, blockchain, cryptocurrency, regtech, and insurtech sectors. Join Canada's Fintech & Funding Community today FREE! Or become a contributing member and get perks. For more information, please visit: www.ncfacanada.org
 |  |  |
|---|---|---|
 |  |  |
Leave a Reply