Lecture 4: Flipped — Agentic AI & LLMs in Finance
Group presentations · token allocation · discussion
4.1 Course objectives
- 4.1 Course objectives
- 4.2 Recap: Agentic AI & LLMs in Finance
- 4.3 Session agenda
- 4.5 Token allocation
- 4.6 Lecturer reflection
- 4.7 Wrap-up & next steps
Welcome to Emerging Technology & Finance
- This is a flipped-classroom Bachelor course: every regular lecture (odd weeks) is followed by a flipped session (even weeks) where all groups present on the same topic. There is no exam to register for — sign up on the course Moodle page by 15 October 2026 so you receive announcements and the token-allocation quiz links.
- Form a group of 4 by the end of Week 1 (Moodle sign-up sheet). Stragglers will be allocated by the lecturers.
- Grading is 100% cumulative across the 6 flipped sessions: each session = 50% peer-allocated tokens + 50% lecturer evaluation. Each group gets 20 fresh tokens every flipped week to allocate to other groups via a Moodle quiz within 5 minutes of the session ending.
- Submission per session: upload your slide PDF to Moodle before each flipped session starts. Ask questions during or right after each session — that is the preferred channel.
- Admin / studies / exam-eligibility questions go to the registrar’s office (Studiensekretariat) at studiensekretariat@uni-ulm.de.
- Course-content questions outside class: email oliver.padmaperuma@uni-ulm.de, CC andre.guettler@uni-ulm.de.
- We also recommend the student advisory service.
Course Objective
Scope
We will:
- Survey six emerging-technology modules at the cutting edge of finance: agentic AI · blockchain & DeFi · fintech business models · RegTech & cybersecurity · CBDCs
- Pair every regular lecture with a flipped session in which every group presents their angle on the topic
- Train critical evaluation, presentation, and peer-judgment skills via a transparent token-based peer-grading mechanic
- Place the technologies in a real-world business and regulatory context (PSD2/3, MiCA, EU AI Act, post-quantum standards)
We will NOT:
- Build production-grade fintech systems or trade live capital
- Cover deep technical implementations (we treat code as supplement, not core)
- Run a separate written exam or final-pitch competition — the cumulative flipped-session grade is the entire grade
Approach
Flipped-classroom alternation (12 weeks)
- Odd weeks (W1, W3, W5, W7, W9, W11): regular lecture introducing the topic
- Even weeks (W2, W4, W6, W8, W10, W12): flipped session — all groups present and allocate tokens
- Groups of 4, formed by end of Week 1
Token mechanic (the grading vehicle)
- 20 tokens per group per flipped session
- Each group allocates them to other groups, weighing insight · originality · clarity · critical depth
- Cumulative across 6 sessions = 50% of final grade · lecturer evaluation = 50%
Course at a glance (1/3)
Foundations of Digital Disruption in Financial Services
What is ‘emerging tech in finance’, how did we get here, where is it going
- Three waves of digital disruption in finance
- Today’s actors: incumbents, challengers, Big Tech, infrastructure
- Regulatory backdrop: PSD2, MiCA, EU AI Act
- Why now: structural drivers
- What this course will cover
Flipped — Digital Disruption in Financial Services
Group presentations · token allocation · discussion
- Recap of the foundations lecture
- Group presentations on digital disruption
- Token allocation & next steps
Agentic AI & LLMs in Finance
From LLMs to agents · applications · failure modes · EU AI Act
- LLMs in finance: architecture, training, capabilities
- Agentic AI: from answers to actions
- Applications: RAG, robo-advisors, AML, trading agents
- Failure modes: hallucination, drift, prompt injection
- Governance: EU AI Act and high-risk obligations
Flipped — Agentic AI & LLMs in Finance
Group presentations · token allocation · discussion
- Recap of the agentic AI lecture
- Group presentations on real LLM and agent deployments
- Token allocation & next steps
Blockchain, Crypto, DeFi & Tokenisation
From distributed ledgers to MiCA-regulated markets
- Blockchain primer: ledgers, consensus, smart contracts
- Crypto markets: BTC, ETH, stablecoins
- DeFi primitives: AMMs, lending, derivatives
- Tokenisation of real-world assets
- MiCA framework and EU enforcement
Course at a glance (2/3)
Flipped — Blockchain, Crypto, DeFi & Tokenisation
Group presentations · token allocation · discussion
- Recap of the blockchain & DeFi lecture
- Group presentations on real protocols and deployments
- Token allocation & next steps
Fintech Business Models
Neobanks, embedded finance, BNPL, Open Banking, Big Tech in finance
- Neobanks: N26, Revolut, Monzo, Chime
- Embedded finance & BaaS
- BNPL: Klarna, Affirm, regulatory pushback
- Open Banking & PSD2 outcomes
- Big Tech in finance
Flipped — Fintech Business Models
Neobanks, embedded finance, BNPL · group presentations · token allocation
- Recap of the fintech business-models lecture
- Group presentations on real companies and unit economics
- Token allocation & next steps
RegTech, Cybersecurity & Privacy-Preserving Compute
Industrialising compliance · cyber-threat landscape · ZKPs, MPC, federated learning · post-quantum
- RegTech overview: industrialising compliance
- KYC/AML automation in production
- Cybersecurity threats in finance
- Privacy-preserving compute: ZKPs, MPC, federated learning
- Post-quantum cryptography & the migration
Flipped — RegTech, Cybersecurity & Privacy-Preserving Compute
Group presentations · token allocation · discussion
- Recap of the RegTech & security lecture
- Group presentations on vendors, incidents, and emerging tech
- Token allocation & next steps
Course at a glance (3/3)
CBDCs & the Future of Money
Wholesale vs retail design · Digital Euro · e-CNY · programmable money
- What’s a CBDC: wholesale vs retail
- The Digital Euro state of play
- China’s e-CNY and small-country implementations
- Programmable money: feature, threat, or both
- Stablecoins as private money: tension with CBDCs
Flipped — CBDCs & the Future of Money
Final session · group presentations · token allocation · course wrap-up
- Recap of the CBDCs lecture
- Group presentations on real CBDC projects
- Token allocation, final standings, and course retrospective
Assignments / Exams
Flipped-Classroom Presentation Series 100% of your grade
Six in-class group presentations across six emerging-tech topics, graded cumulatively. Each session: 50% peer-allocated tokens + 50% lecturer evaluation.
Group of up to 4.
Submit by emailing oliver.padmaperuma@uni-ulm.de, CC andre.guettler@uni-ulm.de. Subject pattern: Emerging Technology & Finance_assignment-1-flipped-classroom-presentations_surname1_surname2_…
21 January 2027
4.2 Recap: Agentic AI & LLMs in Finance
- 4.1 Course objectives
- 4.2 Recap: Agentic AI & LLMs in Finance
- 4.3 Session agenda
- 4.5 Token allocation
- 4.6 Lecturer reflection
- 4.7 Wrap-up & next steps
What we covered last week
- LLMs in finance — the architecture (transformers), training (RLHF), capabilities (summarisation, extraction, reasoning), and the family of finance-specific variants (Chen, Yang, and Liu 2023).
- Agentic AI — moving from “answer a question” to “execute a multi-step task with tool use, memory, and planning.” Why this changes the deployment economics dramatically (Agrawal, Gans, and Goldfarb 2022).
- Concrete applications — RAG over filings and research, robo-advisors, AML/KYC screening, customer-support agents, and the most-claimed but least-deployed: autonomous trading.
- Governance & EU AI Act — risk-based regulation, “high-risk” categories, and the 2026 obligations finance firms face (European Parliament and Council 2024).
- Failure modes — hallucination, prompt injection, distribution shift, model-error catastrophe at scale (Acemoglu and Restrepo 2020).
Notes
The pivot last week was from “AI as a feature” to “AI as a worker.” Agentic systems open a different cost curve: they can do work that previously required headcount, but they also fail in ways no human would. Your job tonight is to find a specific deployment where this trade-off is observable — where you can point to what it replaced, how it failed, and who is now responsible when it does.
Open questions to dig into today
- When does an LLM cross from “smart autocomplete” to “agentic enough to matter”? Is BloombergGPT agentic? Is J.P. Morgan’s IndexGPT?
- Hallucination in finance: insurmountable barrier or solvable engineering problem?
- Robo-advisors have existed for 15 years — why is their AUM still small relative to incumbent wealth managers?
- What does the EU AI Act actually require of a bank deploying an LLM in 2026?
Notes
Don’t accept vendor claims at face value tonight. A bank says it “uses AI”; a vendor says its model is “production-grade.” Ask: what does the model do that was previously done by whom, and what happens when it gets it wrong?
4.3 Session agenda
- 4.1 Course objectives
- 4.2 Recap: Agentic AI & LLMs in Finance
- 4.3 Session agenda
- 4.5 Token allocation
- 4.6 Lecturer reflection
- 4.7 Wrap-up & next steps
Today’s flow
- 14:00 — Welcome & recap (5 min)
- 14:05 — Group presentations (6 min + 2 min Q&A each)
- 15:00 — Lecturer reflection (10 min)
- 15:10 — Token allocation (5 min on Moodle)
- 15:15 — Wrap-up & prep for next regular lecture (5 min)
Notes
Same format as Week 2. Time-box is firm.
Group presentation order
- Angle: A specific deployed LLM in finance (BloombergGPT · FinGPT · IndexGPT)
- Slot: 14:05–14:13
- Angle: A robo-advisor (Betterment · Wealthfront · Scalable Capital) — model and unit economics
- Slot: 14:13–14:21
- Angle: An AI trading agent — real or hype?
- Slot: 14:21–14:29
- Angle: AML/KYC automation case (ComplyAdvantage · Chainalysis · J.P. Morgan)
- Slot: 14:29–14:37
- Angle: EU AI Act × finance — what changes by 2026?
- Slot: 14:37–14:45
- Angle: Catastrophic-risk angle — what’s the worst plausible AI failure in finance?
- Slot: 14:45–14:53
What groups present today
6 min + 2 min Q&A. Each presentation must include:
- One specific deployment (named company, named model, named product).
- What it actually does — the input, the output, the human-in-the-loop role.
- The failure mode you found — not the brochure version.
- 1–2 discussion prompts.
4.4 Group presentations
Group 1 — A deployed LLM in finance
- Candidates: BloombergGPT (proprietary, 50B params), FinGPT (open-source), J.P. Morgan IndexGPT, Goldman Sachs’ “GS-AI Platform.”
- Bring: what task it does, what evaluation benchmark it’s measured on, and what a human used to do that it now does.
Notes
Live-fill slide.
Group 2 — Robo-advisor analysis
- Candidates: Betterment, Wealthfront, Scalable Capital, Nutmeg, Vanguard PAS.
- Bring: AUM, customer-acquisition cost, fee structure, and the structural reason their AUM is still much smaller than incumbents’.
Notes
Live-fill slide.
Group 3 — AI trading agent
- Candidates: Renaissance / Two Sigma claims, retail “AI trading” products, “agentic” hedge-fund pitches.
- Bring: separate the claim from the evidence. If the evidence is opaque, that’s the presentation.
Notes
Live-fill slide.
Group 4 — AML/KYC automation
- Candidates: ComplyAdvantage, Chainalysis Reactor, Feedzai, J.P. Morgan’s COIN, HSBC’s AI screening.
- Bring: a measured false-positive / false-negative trade-off, and what happens when the AI flags wrongly.
Notes
Live-fill slide.
Group 5 — EU AI Act × finance
- Bring: the high-risk category definitions, which finance-AI use cases land there, and the concrete obligations (documentation, human oversight, post-market monitoring) by what date.
Notes
Live-fill slide.
Group 6 — Catastrophic AI failure scenario
- Candidates: a plausible flash-crash-by-LLM scenario; a coordinated prompt-injection at scale; an AML model that systematically under-flags one customer segment.
- Bring: the mechanism, the realistic blast radius, and what regulator / market-structure change would prevent it.
Notes
Live-fill slide.
4.5 Token allocation
- 4.1 Course objectives
- 4.2 Recap: Agentic AI & LLMs in Finance
- 4.3 Session agenda
- 4.5 Token allocation
- 4.6 Lecturer reflection
- 4.7 Wrap-up & next steps
How tokens work
- 20 fresh tokens per group per session.
- Distribute among other groups — no self-allocation. Total = 20.
- Moodle quiz opens for 5 minutes after the last presentation.
Notes
Same rules every flipped week. The 5-minute window is firm.
Allocation rubric
- Insight — taught you something new?
- Originality — fresh angle or the obvious one?
- Clarity — could you follow the argument?
- Critical depth — honest appraisal or sales pitch?
Notes
Tonight especially, reward groups that pushed back on a vendor’s claim — that’s the rarest and most valuable skill in this topic.
Tonight’s leaderboard
| Group | Tokens this week | Cumulative |
|---|---|---|
| Group 1 | — | — |
| Group 2 | — | — |
| Group 3 | — | — |
| Group 4 | — | — |
| Group 5 | — | — |
| Group 6 | — | — |
- Filled in after the Moodle quiz closes.
- “Cumulative” carries forward from Week 2.
4.6 Lecturer reflection
- 4.1 Course objectives
- 4.2 Recap: Agentic AI & LLMs in Finance
- 4.3 Session agenda
- 4.5 Token allocation
- 4.6 Lecturer reflection
- 4.7 Wrap-up & next steps
Standout insights & common gaps
- Standouts (live-fill): 2–3 groups whose evidence on a real deployment was particularly strong.
- Common gaps (live-fill): typically — (a) confused “uses AI” with “is agentic”; (b) failed to surface a measurable failure mode; (c) conflated robo-advisor with autonomous trader.
Notes
The single most common failure mode in Week-4 presentations across institutions is taking marketing language at face value. If a vendor says “our agent autonomously executes trades,” ask who pre-authorised the trade and at what time. The answers are almost always heavily human-in-the-loop.
Concepts to revisit next regular lecture
- The transition from agentic-AI (off-chain, opaque models) to on-chain, transparent code execution is the topic of Lecture 5 — smart contracts as a different kind of “agent.”
- Trust + auditability — AI offers neither natively; blockchain offers transparent state but no semantic guarantee. Where do they meet?
Notes
This bridges to Lecture 5 directly: agentic AI promises action with weak transparency; smart contracts promise transparency with weak semantic flexibility. The intersection is where the most interesting failure modes live.
4.7 Wrap-up & next steps
- 4.1 Course objectives
- 4.2 Recap: Agentic AI & LLMs in Finance
- 4.3 Session agenda
- 4.5 Token allocation
- 4.6 Lecturer reflection
- 4.7 Wrap-up & next steps
Course at a glance (1/3)
Foundations of Digital Disruption in Financial Services
What is ‘emerging tech in finance’, how did we get here, where is it going
- Three waves of digital disruption in finance
- Today’s actors: incumbents, challengers, Big Tech, infrastructure
- Regulatory backdrop: PSD2, MiCA, EU AI Act
- Why now: structural drivers
- What this course will cover
Flipped — Digital Disruption in Financial Services
Group presentations · token allocation · discussion
- Recap of the foundations lecture
- Group presentations on digital disruption
- Token allocation & next steps
Agentic AI & LLMs in Finance
From LLMs to agents · applications · failure modes · EU AI Act
- LLMs in finance: architecture, training, capabilities
- Agentic AI: from answers to actions
- Applications: RAG, robo-advisors, AML, trading agents
- Failure modes: hallucination, drift, prompt injection
- Governance: EU AI Act and high-risk obligations
Flipped — Agentic AI & LLMs in Finance
Group presentations · token allocation · discussion
- Recap of the agentic AI lecture
- Group presentations on real LLM and agent deployments
- Token allocation & next steps
Blockchain, Crypto, DeFi & Tokenisation
From distributed ledgers to MiCA-regulated markets
- Blockchain primer: ledgers, consensus, smart contracts
- Crypto markets: BTC, ETH, stablecoins
- DeFi primitives: AMMs, lending, derivatives
- Tokenisation of real-world assets
- MiCA framework and EU enforcement
Course at a glance (2/3)
Flipped — Blockchain, Crypto, DeFi & Tokenisation
Group presentations · token allocation · discussion
- Recap of the blockchain & DeFi lecture
- Group presentations on real protocols and deployments
- Token allocation & next steps
Fintech Business Models
Neobanks, embedded finance, BNPL, Open Banking, Big Tech in finance
- Neobanks: N26, Revolut, Monzo, Chime
- Embedded finance & BaaS
- BNPL: Klarna, Affirm, regulatory pushback
- Open Banking & PSD2 outcomes
- Big Tech in finance
Flipped — Fintech Business Models
Neobanks, embedded finance, BNPL · group presentations · token allocation
- Recap of the fintech business-models lecture
- Group presentations on real companies and unit economics
- Token allocation & next steps
RegTech, Cybersecurity & Privacy-Preserving Compute
Industrialising compliance · cyber-threat landscape · ZKPs, MPC, federated learning · post-quantum
- RegTech overview: industrialising compliance
- KYC/AML automation in production
- Cybersecurity threats in finance
- Privacy-preserving compute: ZKPs, MPC, federated learning
- Post-quantum cryptography & the migration
Flipped — RegTech, Cybersecurity & Privacy-Preserving Compute
Group presentations · token allocation · discussion
- Recap of the RegTech & security lecture
- Group presentations on vendors, incidents, and emerging tech
- Token allocation & next steps
Course at a glance (3/3)
CBDCs & the Future of Money
Wholesale vs retail design · Digital Euro · e-CNY · programmable money
- What’s a CBDC: wholesale vs retail
- The Digital Euro state of play
- China’s e-CNY and small-country implementations
- Programmable money: feature, threat, or both
- Stablecoins as private money: tension with CBDCs
Flipped — CBDCs & the Future of Money
Final session · group presentations · token allocation · course wrap-up
- Recap of the CBDCs lecture
- Group presentations on real CBDC projects
- Token allocation, final standings, and course retrospective
Prepare for next regular lecture
- Skim the original Bitcoin whitepaper (Nakamoto 2008) and the Ethereum whitepaper (Buterin 2014) (sections 1–3 of each).
- Track one DeFi event this week: a hack, a launch, a regulatory action, a tokenisation deal.
- Group discussion Sunday: pick a Week-6 angle on Blockchain / DeFi / Tokenisation.
Notes
Lecture 5 covers a lot of ground; pre-reading the whitepapers (they are short) lifts the lecture by an order of magnitude.
See you next time
- Token allocation: Moodle quiz open now — 5 minutes.
- Slide PDFs from all groups archived under Week 4.
- Lecture 5 (next Thursday): Blockchain, Crypto, DeFi & Tokenisation — what’s actually being built, what’s smoke.