Couverture de Won't Fix

Won't Fix

Won't Fix

De : Rob Leathern
Écouter gratuitement

Offres de saison | 0,99 € par mois pendant les 3 premiers mois

5,99 €/mois par la suite. Des conditions s’appliquent. Annulation mensuelle possible.
From the founders of InfoHawk: conversations about AI-driven deception, abuse and scams, and why they’re so hard to stop. In software engineering, “won’t fix” describes a bug by acknowledging the issue but intentionally leaving it unsolved because addressing it is too costly, risky, or not worth the trade-offs. Hear from the practitioners fighting phishing, deepfakes and bots, and learn about the broken systems and misaligned incentives that keep us all vulnerable.Copyright 2026 Rob Leathern Direction Economie Management et direction Politique et gouvernement
Épisodes
  • Won't Fix Episode 18: With Jon Leibowitz Former Chairman, FTC

    Won't Fix Episode 18: With Jon Leibowitz Former Chairman, FTC
    Oct 9 2026
    My guest Jon Leibowitz spent nearly a decade at the Federal Trade Commission, and four years running it — as Chairman from 2009 to 2013, right through the wreckage of the financial crisis.During Jon’s chairmanship (March 2009 to February 2013), the FTC's enforcement centered on "last dollar" frauds preying on Recession-battered consumers: coordinated national sweeps against mortgage-rescue and foreclosure-relief scams (leading to the 2010 MARS Rule banning advance fees), and a landmark $108 million settlement with Countrywide over mishandling struggling borrowers. It hit deceptive marketers hard — the roughly $359 million Jesse Willms "free trial" negative-option case, a $163 million scareware judgment against fake-virus-pop-up operators, the 2011 fake-news-site sweep over bogus acai-berry weight-loss claims and phony Oprah and Rachael Ray endorsements, and a 2012 international crackdown on India-based tech-support scam boiler rooms.On the robocall front, it brought cases like the Cash Grant Institute and the auto-warranty scheme (the telemarketer who surrendered his Mercedes) and launched the public Robocall Challenge contest. Beyond fraud, his FTC also built the agency's modern privacy agenda — landmark settlements with Google and Facebook and a record $22.5 million penalty against Google over Safari tracking — and stayed active on antitrust and competition alongside all of it.After the FTC, he went into private practice at Davis Polk, testified before Congress more times than he can probably count, and today he chairs the National Consumers League — one of the oldest consumer advocacy groups in the country.But here's what makes this conversation timely. The scams he chased fifteen years ago never died. They got faster, cheaper, and — thanks to AI — a whole lot more convincing.Chapter Timestamps:00:00 Intro2:33 FTC career and current work4:03 Old scams, new technology, and the rise of AI-enabled fraud5:30 From the “spam king” to large-scale online deception6:51 Fraud supply chains, overseas actors, and cross-border enforcement9:52 AI deepfakes and the persuasive power of modern scams11:21 Affiliate marketing, robocalls, and weakened FTC monetary remedies14:42 State attorneys general and platform cooperation17:48 - Shared intelligence and the multi-provider nature of scams18:23 AI fraud’s scale and fabricated “evidence”20:33 Long-form fake media and recurring-subscription traps23:39 Domain authentication and the failure of voluntary privacy standards27:03 Reporting, independent evaluation, and regulatory balance33:02 Trust in the internet and FTC rules against impersonation and fake reviews34:53 Business verification and the trade-off between friction and safety37:14 A $0 trial that becomes $34.90 per month39:58 Closing: protecting reality while preserving innovationResources & Links:Rob Leathern: (https://www.linkedin.com/in/leathern/)Jon Leibowitz: (https://www.linkedin.com/in/jon-leibowitz-14091679/)Steve Martin scam video website: https://assist.saeculums.com/ Cases from Jon's era:Sanford Wallace "Spam King" (CD-ROM tray) — FTC release: https://search.ftc.gov/news-events/news/press-releases/2004/10/ftc-cracks-down-spyware-operation — The FTC's 2004 complaint describes spyware that popped open CD-ROM trays and then flashed a "FINAL WARNING!!" pushing paid fixes. Nice follow-up detail: a court later ordered Wallace to give up $4,089,500, from selling the "fix" at roughly $30 a copy.Scareware $163M judgment (Kristy Ross): https://www.ftc.gov/node/47528 — Useful context for listeners: a co-defendant and his father had to give up $8.2 million under a 2011 settlement, which is a good contrast with the giant headline number.2012 international tech-support crackdown: https://www.ftc.gov/news-events/news/press-releases/2012/10/ftc-halts-massive-tech-support-scams — This is the press conference Jon describes; one of the six operations found victims by buying Google ads that appeared when people searched for their computer company's support number, which echoes your toll-free-number point.Jesse Willms $359M settlement: https://www.ftc.gov/news-events/news/press-releases/2012/02/ftc-halts-deceptive-practices-marketer-who-collected-359-million-using-bogus-free-trial-offersThe Atlantic, "The Dark Lord of the Internet" (Willms profile, 2014): https://www.theatlantic.com/magazine/archive/2014/01/the-dark-lord-of-the-internet/355726/Fake news sites / acai sweep (2011): https://www.ftc.gov/news-events/news/press-releases/2011/04/ftc-seeks-halt-10-operators-fake-news-sites-making-deceptive-claims-about-acai-berry-weight-lossCentral Coast Nutraceuticals (fake Oprah / Rachael Ray endorsements): https://www.ftc.gov/news-events/news/press-releases/2012/01/internet-marketers-acai-berry-weight-loss-pills-colon-cleansers-pay-15-million-settle-ftc-chargesAMG Capital Management v. FTC (SCOTUSblog case page): https://www.scotusblog.com/case-files/cases/...
    Afficher plus Afficher moins
    41 min
  • Won't Fix Episode 17: With Dave Willner, Co-Founder of Zentropi

    Won't Fix Episode 17: With Dave Willner, Co-Founder of Zentropi
    Sep 25 2026

    My guest today is Dave Willner, cofounder of Zentropi and a fixture in the Trust and Safety world.

    In 2008, he was a young Facebook employee who sat down to replace a one-page list of banned content — it said, roughly, no Hitler and no naked people — with the first real set of community standards the internet had ever seen. The rules he wrote are still the foundation of how Meta governs content shared by billions of people.

    He went on to build community policy at Airbnb, and then joined OpenAI in early 2022, before ChatGPT existed, to figure out how to stop an image generator from producing bad things.

    Now he's co-founded a company called Zentropi, with Samidh Chakrabarti, which is built on the idea that moderation at scale was never actually impossible — we just didn't have the right tools.

    In This Episode:

    Human moderation suffers from high turnover, trauma, and inconsistent judgment—problems small, tailored language models solve by strictly enforcing written rules.

    Specialized small models can run circles around massive frontier models on speed and cost, making real-time classification financially viable.

    Written policies and labeled datasets are two sides of the same coin; when an AI disagrees with a human label, it usually exposes a vague policy rather than a model error. Running moderation models locally on your own servers cuts per-check costs, dodges European privacy paperwork, and keeps user data in-house.

    Bad actors are using AI to scale exploits at high speeds, forcing safety teams to adopt AI defenses immediately just to keep up.

    A single global moderation rule doesn't work—dating apps, teen platforms, and social feeds require custom enforcement thresholds based on their specific risks.

    Using the same LLM to draft a policy, generate test cases, and grade itself creates lazy evaluation loops that hide major safety blind spots.

    Giving open-weights AI tools to volunteer moderators on Reddit or Discord lets small teams govern massive communities without burning out.

    Chapter Timestamps:

    00:00 Intro

    3:21 Dave Willner’s trust-and-safety background and the origins of Zentropi

    7:04 Zentropi’s specialized moderation models and policy-optimization tools

    11:15 Why human moderation is inconsistent and difficult to measure

    15:45 The limits of human judgment and the case for automated moderation

    21:09 Imperfect language, model precision, and faster adaptation to abuse

    24:43 Shared moderation infrastructure versus platform-specific policies

    28:46 Why small specialized models can outperform larger general models operationally

    32:26 Local deployment, privacy, and trust-and-safety resource constraints

    34:16 Customer economics and real-time video moderation

    38:09 The danger of “slop” policies and weak AI evaluations

    44:10 AI moderation as a tool for decentralized community governance

    47:06 The adversarial threat landscape and urgency of AI-enabled defense

    Resources & Links:

    Zentropi (https://zentropi.ai/)

    Dave Willner (https://www.linkedin.com/in/davewillner/)

    Rob Leathern (https://www.linkedin.com/in/leathern/)
    Afficher plus Afficher moins
    53 min
  • Won't Fix Episode 16: With John Canfield, Founder & CEO of BlueArc

    Won't Fix Episode 16: With John Canfield, Founder & CEO of BlueArc
    Sep 17 2026

    The question "is this business real, and should I believe what it's telling me" isn't new. It's about as old as American commerce. What's changed is that you can now spin up a company, a website, and a merchant account before lunch, and there's no correspondent in town to write you up.

    My guest has spent his career on the modern version of that problem. John Canfield built risk and verification systems at eBay, at WePay (acquired by JPMorgan Chase) and at Google, where he led the Ads verification and transparency initiative. He's now co-founder and CEO of BlueArc, which uses AI to verify business customers.

    Some themes:

    1. How verification became paperwork The Bank Secrecy Act, then post-9/11 customer identification rules, hard-coded KYC as document collection — optimized for what an examiner could inspect, not what would catch a bad actor. When did "compliant" and "works" come apart?

    2. Identity versus credibility He led Google Ads verification in 2020 and now argues that approach won't stop scams: identity answers who's speaking, credibility answers whether to believe them. Platforms spent a decade saying claim-level review couldn't scale, so why now?

    3. Flipping the economics — who actually pays If the goal is no added friction for legitimate businesses, and if platforms can charge for deeper review, what then?

    4. FTC v. Genesis Tech (filed June 2026, N.D. Cal.): we talked about this case, where the FTC alleges a network of 15 corporations and 8 individuals ran deceptive subscription apps through a shifting web of Cyprus and Delaware shell companies, continually registering new entities and merchant accounts to outrun fraud monitoring — with linked PayPal accounts processing close to $700M in the twelve months ending September 2025. Per the FTC’s standard, a practice is deceptive if it's likely to mislead a consumer acting reasonably in the circumstances and is material to their decision — no intent and no actual victims required, and net impression governs, so an ad that is literally true in every sentence can still be unlawful.

    Chapter Timestamps:

    00:00 Introduction

    1:21 Jon Canfield’s trust-and-safety background and BlueArc’s mission

    3:34 Why business verification differs from verifying individuals

    5:36 The missing business graph behind platform risk decisions

    7:20 Legal entities are not enough: domains as a business identity layer

    11:32 Balancing low-friction verification with domain control and vouching

    16:36 Verification does not equal credible advertising claims

    19:08 Risk-based claim credibility, transparency, and the limits of black-box enforcement

    23:24 AI-enabled advertiser engagement and remediation

    27:00 Proportional friction: local flower ads versus high-risk miracle claims

    30:56 Scam economics and why advertisers may need to fund deep validation

    34:16 Shared third-party validation and BlueArc’s next priorities

    39:19 FTC material misrepresentation as a scalable policy framework

    42:38 Know Your Agent: AI shopping agents, authorization, and scam resilience

    46:02 Review-site integrity and adversarial manipulation

    Resources & Links:

    Rob Leathern (https://www.linkedin.com/in/leathern/)

    John Canfield (https://www.linkedin.com/in/johncanfieldbayarea/)

    BlueArc (https://bluearc.ai/)
    Afficher plus Afficher moins
    49 min
adbl_web_anon_alc_button_suppression_t1
Aucun commentaire pour le moment