The AI Argument - EP56 Dead Man Speaks in Court, Bot Dupes 46K on X, and AI Act Tensions

The AI Argument

The EU wants to lead the world on trustworthy AI, but can it really regulate its way to the front? Frank is optimistic. Justin rolls his eyes. What starts as a polite difference of opinion quickly turns into a pointed question: is the EU building the future, or tying it up in red tape?
Frank backs the EU AI Act as a serious attempt to set global standards, pointing to its ambition and echoes of GDPR’s success. Justin sees a different story, regulation slowing Europe’s progress, while the US and China charge ahead, unbothered by Brussels’ good intentions. For him, this isn’t about compliance, it’s about whether Europe can stay relevant in a race fuelled by code, not policy.
If you’re trying to stay ahead of AI, or at least not get run over by it, this is exactly the kind of friction worth paying attention to.
From there, things don’t get any calmer. Justin declares hallucinations solved. Frank’s not having it. They argue over OpenAI’s $3B Windsurf acquisition (is AGI closer or further than it looks?), Stripe’s incredible fraud detection AI, and a court case where an AI avatar spoke for a murder victim.
And just when you think things can’t get weirder, Justin confesses he got fooled by an AI bot on Twitter. A smart one. With opinions.

챕터

1. Can EU regulation really spark innovation? (00:00:00)

2. Is hallucination just bad engineering? (00:12:14)

3. Why did OpenAI drop $3B on Windsurf AI? (00:17:54)

4. Will AI agents soon shop with your credit card? (00:22:17)

5. Is Stripe’s AI insanely good at spotting fraud? (00:23:34)

6. Can an AI speak for the dead in court? (00:25:18)

7. Do we need Blade Runners for Twitter now? (00:32:20)