Six Weeks Away: The AI Law Most Companies Don't Know They're Breaking

Six Weeks Away: The AI Law Most Companies Don't Know They're Breaking

Reflect w/ Ed Fassio — where AI tells the stories that matter. August 2, 2026 is the enforcement date for the EU AI Act's high-risk AI provisions — and most enterprises still don't know if they're covered. HR tools, hiring algorithms, credit scoring systems, employee monitoring software — all potentially in scope. Penalties go up to €15 million or 3% of global annual turnover. This isn't a future regulation. It's here. Your Move: If you're deploying AI that touches hiring, credit, benefits, or worker monitoring, you need a compliance audit now. Start with an AI inventory — most organizations don't even have one. Treat any ambiguous system as high-risk until you can prove otherwise. — Reflect w/ Ed Fassio | reflectpodcast.com Send us Fan Mail [https://www.buzzsprout.com/2414886/fan_mail/new] Support the show [https://www.buzzsprout.com/2414886/support] LISTEN TO MORE EPISODES: https://www.reflectpodcast.com [https://www.reflectpodcast.com]

The US, the EU, and China Are Writing Three Different Futures. Only One Wins.

Reflect w/ Ed Fassio — where AI tells the stories that matter. The US, the EU, and China are writing three fundamentally different answers to the same question: how should AI be governed? And whichever answer wins — or gets adopted by enough of the world's economies — becomes the default for your business, your data, your workforce, and your competitive position. Whether you had a vote in it or not. Innovation-first. Rights-first. State-first. Three philosophies. Not converging. Julius and Hale break down what each model means in practice, why the Brussels Effect means the EU standard will likely become the de facto global baseline, and what that means for every organization deploying AI today. This is the final episode of The Governance Question arc. Four episodes. One spine: Aza Raskin's warning that AI doesn't need to be malicious to be dangerous. Indifference is enough. Governance is how humanity stays present in its own future. Your Move: Know your regulatory exposure. Adopt the most demanding standard you face as your org-wide baseline. And engage — the policy conversations happening right now will shape the frameworks you operate under for the next decade. — Reflect w/ Ed Fassio | reflectpodcast.com Send us Fan Mail [https://www.buzzsprout.com/2414886/fan_mail/new] Support the show [https://www.buzzsprout.com/2414886/support] LISTEN TO MORE EPISODES: https://www.reflectpodcast.com [https://www.reflectpodcast.com]

Every Industry Has a Safety Standard. AI Doesn't. Here's What One Would Look Like.

Reflect w/ Ed Fassio — where AI tells the stories that matter. When a new drug goes to market, it clears a decade of trials. When a bridge gets built, every spec gets certified. When a financial product launches, there are disclosure requirements and audit trails. Every high-stakes system humanity has built has a safety standard. Except AI. A company can buy a model, connect it to their HR workflow, and start making decisions about people's livelihoods with less regulatory oversight than a food truck needs to sell tacos. Julius and Hale lay out what a minimum viable governance baseline actually looks like — five components any organization can implement without a dedicated compliance team. Not the EU AI Act. Not ISO 42001. The Monday morning version. The five: Model Card. Purpose Boundary Document. Audit Trail. Human Checkpoint Map. Incident Response Protocol. Your Move: Pick one AI system your org is currently using that touches a decision about a person. Run the five-component check. Count your yes answers. That's your governance score. Anything below five is a gap worth closing this quarter. — Reflect w/ Ed Fassio | reflectpodcast.com Send us Fan Mail [https://www.buzzsprout.com/2414886/fan_mail/new] Support the show [https://www.buzzsprout.com/2414886/support] LISTEN TO MORE EPISODES: https://www.reflectpodcast.com [https://www.reflectpodcast.com]

AI Literacy Is Job Security Now: The 3x Layoff Gap Nobody's Talking About

Gallup just released data that tech leaders need to hear. Workers who use AI at least monthly face a 6% layoff risk. Workers who don't? 18%. That's a three-fold difference. And it's not an accident — it's a signal about who survives the next wave of industry restructuring. Send us Fan Mail [https://www.buzzsprout.com/2414886/fan_mail/new] Support the show [https://www.buzzsprout.com/2414886/support] LISTEN TO MORE EPISODES: https://www.reflectpodcast.com [https://www.reflectpodcast.com]

The Algorithm Decided. Nobody's Responsible. That's Not an Accident.

Reflect w/ Ed Fassio — where AI tells the stories that matter. Robert Williams was arrested in Detroit — handcuffed in front of his daughters — because a facial recognition AI misidentified him. 900 UK Post Office managers were prosecuted over 14 years because a faulty software system generated false shortfalls. 50,000 patients were denied extra care because a health algorithm used cost as a proxy for need. Three different systems. Three different kinds of harm. And in every single case — nobody was held accountable. This isn't bad luck. It's a structural problem with three layers: the liability gap, the transparency gap, and the audit gap. Julius and Hale break down how accountability dissolves — and what an accountability map actually looks like before you need one. Your Move: Run five questions on your most critical AI deployment right now. What decisions does it influence? Worst plausible outcome? Named human accountable? Escalation path? Contestability mechanism? Count your yes answers. That number is your governance score. — Reflect w/ Ed Fassio | reflectpodcast.com Send us Fan Mail [https://www.buzzsprout.com/2414886/fan_mail/new] Support the show [https://www.buzzsprout.com/2414886/support] LISTEN TO MORE EPISODES: https://www.reflectpodcast.com [https://www.reflectpodcast.com]