In January 2026, Google was granted a patent that lets it read your website, break it into components, and serve searchers an AI-generated page assembled from your content. Visitors may never see your actual site. Joe Toscano — former Google designer, one of the voices in The Social Dilemma, and founder of Service Stories — saw this coming.

Joe left Silicon Valley in 2017, before Cambridge Analytica broke, because of what he was being asked to build. He now runs Service Stories, which helps small and mid-size service businesses get found by AI search engines by converting their existing work orders and job data into content — automatically. Early results from their pilots show direct web traffic growth of 228% within 90 days, and over 400% across longer periods.

The shift that makes this possible: search is moving from "plumber near me" to "why is my HVAC system creating complications in my house?" Topical authority — specific, question-answering content — is now the edge, and AI makes it viable to produce at scale for the first time.

He also just returned from two months in China — and his account of robots delivering room service in three minutes, seamless phone-based identity and payments, and a level of operational fluency the West hasn't reached yet is genuinely uncomfortable for anyone who assumes Europe's regulatory caution is the only sensible posture.

A sharp, grounded episode on where search is going, what China is building, and what AGI actually looks like when it arrives.

Find Joe and a step-by-step guide to doing this yourself at servicestories.com.

Hosted on Acast. See acast.com/privacy for more information.

Simon Hodgkins has spent a career running global marketing — Sage, CMO Club president, and now Chief Marketing Officer at Vistatec, the Dublin firm helping iconic brands scale content and AI worldwide. He joins Maryrose to argue that the real AI bottleneck isn't the models. It's whether a company is ready to use them.

Simon makes the case for gap analysis before tooling: before you buy, pilot or deploy, you need to know how you actually work today — what content exists, which workflows are documented, how the Frankfurt office differs from New York, and what ISO 42001 would expect you to prove. He explains why bolting AI onto broken processes wastes money, why regulated industries still need human experts in the loop, and what an "intelligent workflow" actually looks like.

On the hype cycle, Simon is blunt: stop headline-reading. Figure out what you're doing first. Or as Rory Sutherland puts it — no robots are buying new cars.

A conversation for any leader whose pilots keep going nowhere.

Hosted on Acast. See acast.com/privacy for more information.

The CSO stopped tracking the price of a pint of Guinness in Ireland in 2011. The last official figure was €3.93. Nobody picked it up again until Matt Cortland and John Fleming decided to call every pub in the country and ask.

Using Google Maps for contact data, 11 Labs to clone a Northern Irish voice, Claude Code to orchestrate the whole system, and Twilio to actually make the calls, they built the Guinndex: a real-time index of pint prices across Ireland and Northern Ireland. Over 700 pubs called. Total cost: approximately €200.

They didn't stop there. With fuel prices dominating headlines on both sides of the Atlantic, they applied the same methodology to US gas stations, creating the Gas Index, complete with AI agents named after King of the Hill characters and a feature that tells you whether the cheaper station is actually worth the drive.

This episode is a masterclass in what's possible when you connect off-the-shelf AI tools with a problem worth solving. And a cautionary tale about the Irish goodbye.

Find Matt and John at guindex.ai.

Hosted on Acast. See acast.com/privacy for more information.