Dec 16, 2024 · 11m · a16z
When Regulation Becomes Code
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
An a16z partner explains how recent advancements in artificial intelligence allow complex regulatory frameworks to be encoded into software, mitigating massive operational compliance costs and enabling a new wave of fintech innovation. By automating compliance workflows and replacing legacy systems, AI tools empower startups, streamline institutional processes, and ultimately deliver superior financial products to consumers.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
The host as informed peer 2.6 Guest teaching 5.0 Guest disagreement 0.6 The host pushing back 1.0
speaking balance: gold is the host, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 2:27 Playful teasing on leading the witness
In an overall non-combative interview, the guest playfully quizzes the host on job growth statistics and notes she led the witness, marking the closest instance to playful conversational tension.
Hardest push from the host ▶ 4:14 Host challenges LLM capability in gray areasThe host directly asks whether AI products can actually interpret gray areas in compliance, pressing on potential failure modes before the guest explains hallucination controls.
Biggest teaching moment ▶ 1:10 Guest explains modern compliance breakdown and TD Bank fineThe guest thoroughly educates the host on how manual compliance relies on brittle workflow tools and manual web research, resulting in massive regulatory fines.
The host holds their own ▶ 1:57 Host cites granular TD Bank violation detailsThe host demonstrates deep preparation by citing precise data points on TD Bank's unmonitored detection alerts and investigation backlogs.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The host as informed peer | Guest teaching | Guest disagreement | The host pushing back | Why |
|---|---|---|---|---|---|---|
| Big Ideas 2025: When Regulation Becomes Code | 3 | 5 | 1 | 0 | The host demonstrates background knowledge by citing specific figures regarding TD Bank's compliance failure cases. The guest educates the host on the scale of federal banking rules and how manual compliance workflows operate. The dynamic is highly collaborative with no combativeness or host pushback. | |
| The Rise of AI Compliance Copilots | 3 | 6 | 1 | 3 | The host correctly guesses the compliance job growth statistic and questions whether LLMs can genuinely handle gray areas in regulatory compliance. The guest explains how LLMs handle rules like the SEC marketing regulation and addresses hallucination mitigation. The exchange remains friendly while addressing real technical limitations. | |
| Accelerating Market Adoption and Modernizing Systems of Record | 2 | 5 | 1 | 1 | The host prompts the discussion with relevant observations about slow enterprise adoption in banking and insurance. The guest provides market insight into legacy replacement trends like Sardine replacing Actomize and new buyer personas. Both parties maintain a cooperative tone. | |
| Unlocking Long-Term Economic and Consumer Benefits | 3 | 5 | 0 | 1 | The host thoughtfully reframes her own question to focus on economic opportunity costs created by heavy regulatory overhead. The guest elaborates using Robinhood, Chime, and SBA loan processing delays as illustrative examples. The dialogue is smooth and constructive without conflict. | |
| Why Regulation as Code Matters to Everyone | 2 | 4 | 0 | 0 | The host steers the conversation to consumer relevance, asking how regulation as code impacts the everyday person. The guest explains how lower compliance barriers drive fintech innovation and lower consumer costs. The segment concludes on a fully aligned, informational summary. |
Statements from this episode (22)
Strange: Federal agencies enforce 50,000 regulations, up from 10,000 in 1970
“There are 50,000 different rules and regs across different agencies, and that's up from 10,000 in 1970.”
Strange: Regulation will become software code in 2025
“In 20, 25 regulation is going to become code.”
Strange: Regulation is a top cost for banking, healthcare, and insurance
“Regulation is one of the biggest costs to existing companies, banks and insurance companies and healthcare companies specifically, and a huge burden to new entrants coming in and to innovation.”
Strange: US efficiently adds regulation but never repeals old rules
“And I think in this country, we have a system where we efficiently add regulation, but we never take away regulation.”
Angela Strange: State regulatory growth mirrors federal banking regulation expansion
“That's just federal. Every state is following exactly the same trend line.”
Strange: TD Bank Fined $3 Billion Over Compliance Alert Backlog
“And most recently, big headline, TD Bank was fined three billion dollars for having a long backlog of these compliance alerts.”
Strange: Up to 15% of large bank staff works in compliance
“So large banks right now, up to 15% of their staff is just in compliance.”
Strange: Compliance is the fourth fastest-growing US job over 20 years
“You want to know the fourth fastest growing job in the United States in the last 20 years? ... Compliance.”
Strange: Every white-collar worker will eventually have an AI copilot
“I believe that every white-collar worker eventually is going to have a copilot, and some of those roles will be replaced by agents.”
Strange: a16z compliance team reviews every tweet for SEC compliance
“Anytime we want to tweet something, we are a registered investment advisor. We have a great compliance team. We have to send our tweets over, and they will tell us if we are compliant with the SEC marketing rule.”
Strange: Norm AI is codifying specific regulations into automated compliance software
“And there's companies, for instance, Norm AI is one that's doing exactly that. It is codifying specific regulation such that you can send it things and it'll come back and tell you the answer.”
Strange: LLM hallucination is a feature in consumer apps, but a bug in compliance
“So LLM's hallucinating, and for my consumer colleagues, that's a feature. It comes out with great, great creative, fun and enhancing things in the world of compliance that is definitively a bug.”
Strange: Successful compliance startups use AI copilots rather than autonomous agents
“Smarter entrepreneurs are getting around this in two different ways. One it's much more of a co-pilot versus a full angel versus a full agent. And then they're able to Test enough edge cases around that the likelihood of hallucination is very, very, very low b…”
Angela Strange: Customers Are Replacing 20-Year-Old Actimize Software with Sardine
“So for instance, Sardine is a company that's got a much more modern transaction monitoring system, and they are seeing customers rip and replace Actomize that's been around for 20 years.”
Angela Strange: AI Compliance Startups Achieving Unusually Fast ACV Growth
“It is moving a lot more quickly than I would have thought. You know, I've been investing in and around the space for the last 10 years, and the momentum and even just the ACVs that some early companies are getting is, is quite fast.”
Angela Strange: AI Compliance Startups Overcome Hallucinations by Fine-Tuning and Showing Work
“I think the two things that come up with the blockers where the smart founders have gotten around are, one, how do I prevent this from hallucinating? And I think the two ways are, one, showing the work, And then two, proving that you have fine-tuned your model…”
Angela Strange: Chief Compliance Officers Emerged as Primary Tech Buyers
“Chief compliance officers, like BSA officers. Like, I imagine all of these All of these potential buyers are now all of a sudden getting a lot of interest from very technical, technical solutions.”
Strange: Robinhood forced zero-commission trading across major U.S. brokerages
“Trading now is free. I would argue that's in large part because Robinhood came along, made trading free, and put pressure on all the large brokerages.”
Strange: Chime expanded access to free checking accounts by pressuring incumbents
“For a large portion of the of the U.S. Population, it was very difficult to get a free checking account. Chime came along, enabled that to be possible, and put pressure on the incumbents.”
Strange: SBA loans take 90 days due to 1,000-page compliance documentation
“Many businesses get loans from the SBA. It takes, on average, 90 days. Why? Part of that is there's a thousand-plus page documentation about how do you make this loan compliant, right?”
Strange: Automating tasks done by dozens of employees provides AI entry wedge
“And then .2 would be any area where a company needs to hire, you know, dozen plus people to do a task where those people would rather do something a little bit more strategic. That can be a very interesting wedge in for a co-pilot type co-pilot type opportunit…”