Why opting out of AI is actually harming your future self with Kate Minogue

The first time ChatGPT turns your chaotic thoughts into something clear and convincing, it can feel like magic. But what, exactly, are we impressed by? Research published in the Journal of Marketing found that people with lower AI literacy were generally more receptive to AI, with feelings of magic and awe helping explain the pattern. In this solo episode, I explore what that finding might tell us about trust, why using AI every day isn’t the same as understanding it, and the human qualities we read into a fluent answer. I also consider why mystique sells, and share four questions that help me judge where AI is useful and where my own judgement needs to stay involved. I’m Ailish McLaughlin, and this is Trust Issues: the show where we read the label on AI so you can decide for yourself what’s worth your time, your money and your trust. Research discussed: “Lower Artificial Intelligence Literacy Predicts Greater AI Receptivity” — Stephanie M. Tully, Chiara Longoni and Gil Appel (2025). https://doi.org/10.1177/00222429251314491 Instagram: https://www.instagram.com/ailishgetsbetter/?hl=en Substack: https://ailishmcl.substack.com/

My dad saw the headline that an OpenAI agent had escaped its test environment and hacked into Hugging Face, and panicked. I saw the same headline and shrugged. But neither reaction told the full story. In this solo episode, I explore how to make sense of AI headlines when we’re either missing the context or drowning in the noise, and share three questions to ask before we panic, dismiss or pass a story on. A note on timing: I recorded this episode before further findings about the Hugging Face incident were published, including more detail about agents communicating and coordinating outside their intended boundaries. Those findings change parts of our understanding of the incident, so this episode reflects what was known when I recorded it. They don’t change its central message: making sense of AI headlines means noticing both the context we’re missing, the noise we’ve become used to, and being willing to update our view as we learn more. I’ll also be putting these questions into practice in a daily Instagram Reels series, Read the Label: AI Headlines, looking at one AI news story at a time. [https://www.instagram.com/ailishgetsbetter/?hl=en] Further reading: [OpenAI’s fuller investigation](https://openai.com/index/hugging-face-incident-and-the-road-ahead/). Read my writing: https://ailishmcl.substack.com/

AI note-takers might be one of the simplest ways to start using AI at work. But once every meeting becomes searchable, queryable context, things get interesting very quickly. In this episode, I speak to Camilla Bier from Granola about why the first AI-generated note can feel like an instant “aha” moment — and what happens when the note-taking tool starts becoming something much bigger. We get into how Granola actually works, what happens to your data when third-party transcription and LLM providers are involved, and why its providers aren’t allowed to train on customer conversations. We also talk about the increasingly blurry etiquette around recording meetings, whether we’re heading towards a world where we simply assume conversations are being captured, and what outsourcing our memory to AI might be doing to our brains. Plus: why Camilla thinks our generation has a responsibility to teach people not just to use AI, but to critique what it gives us.

Looking for a job was already brutal. Now there's AI on both sides of it. You use AI to fire off applications, the company uses AI to screen them, and somewhere in the middle an actual human may never read a word you wrote. So how do you stand out when everyone's using the same tools? This week Ailish sits down with Roxanne Taku and Joana Rocha, co-founders of TechTalk, a career community with over 6,000 members and a track record of getting people hired at the likes of Google, Salesforce and TikTok. They've also built Reslink, a platform that puts a short video of you on top of your CV, with the aim of bringing the human back into hiring. Roxanne fought her way into UK tech from South Africa and came back from the 2023 layoffs with offers from Salesforce and HubSpot. Joana arrived from Brazil needing a visa sponsor and became a marketing leader in fintech. They've lived the job hunt from every side, so they know how it actually works now. Roxanne and Joana walk us through what really happens to your CV after you hit apply. The keyword scoring, the AI screens, the interviews now run by a bot, and the "hidden job market" where the best roles never get posted at all. They're refreshingly practical about what works, and honest about what doesn't. And here's the stat that reframed the whole thing for me. Roxanne reckons 80 to 90 percent of the people in their community who land interviews get there through networking, not cold applications. So the numbers game most of us are playing might be the wrong game entirely. We also get into where this is all heading, because if AI keeps changing work this fast, the job stops being just doing the work. It becomes thinking about how you add value, far earlier than we used to. In this episode, we get into: What really happens to your CV after you apply, from keyword scoring to AI interviews The hidden job market, and why the best roles often never get posted Why networking, not mass applications, is how most people actually get hired How to build a LinkedIn profile that makes recruiters come to you Whose bias gets baked in when an algorithm screens your application The 70 to 80 percent rule, and which bits of a job spec you actually have to hit Using AI to get better at your job, not just faster What all of this means for graduates and anyone starting out

AI makes it easier than ever to avoid the hard bits of thinking. The blank page, the long document, the messy first draft, the uncomfortable five minutes before an idea clicks, AI means we never have to experience any of these anymore. In this solo episode of Trust Issues, Ailish explores what happens when our shortcut-seeking brains meet tools designed to remove friction. From AI cravings and workslop to the risk of outsourcing our judgment, this episode asks: which parts of our thinking are worth keeping? Plus, five practical guardrails for using AI without losing your edge.