Holly and Ewan Are Working On It Podcast

Episode 10: The AI Risks We’re Not Ready For

In this episode, Holly and Ewan deliberately shift tone to explore the risks, power dynamics, and uncomfortable questions surrounding AI, from superintelligence fears to geopolitics, cyber warfare, and who really controls the technology shaping our lives. Ewan opens with a stark provocation from the book If Anyone Builds It, Everyone Dies [https://amzn.eu/d/eBv2hUj], which argues that the creation of superintelligent AI could represent an existential threat to humanity. Holly discusses the concentration of power in the hands of unelected tech leaders and questions how democratic oversight can exist when AI systems increasingly influence behaviour, economies, and national security. Together, they explore why sovereign AI, sovereign cloud, and even sovereign chips are becoming national priorities, from Australia to the Gulf to Europe. This is the 10th episode marking the end of Season 1. Thank you so much for listening and standby for Season 2!

Episode 9: The Year AI Went Mainstream

In this episode, Holly and Ewan look back on 2025 as a defining year in technology, exploring what really changed (and what didn’t) in the most transformative period since generative AI emerged. They begin with the obvious headline: ChatGPT’s rise into the top five most-visited websites on Earth, a staggering shift for a product only three years old. Ewan reflects on the moment AI moved from novelty to normalised daily tool, from homes to boardrooms to government reports. Holly highlights how this year’s explosion of ChatGPT features (apps, search, shopping recommendations, and especially AI voice conversations) have reshaped expectations of what interacting with AI feels like. Ewan shares real-world stories of fixing household appliances using ChatGPT voice guidance, contrasting it with Google’s Gemini, which has undergone a dramatic improvement arc. The pair also discuss: * Email-integrated AI (e.g., Gmail → ChatGPT via Pulse) * The coming wave of agentic automation, and why agents still aren’t truly mainstream * How Gemini 3 and Google’s AI reboot shocked the industry * Nvidia’s rise to the world’s most valuable company * Why predictions of mass AI-driven unemployment haven’t materialised (yet!) * The uncomfortable truth: companies use “AI realignment” as a convenient narrative during layoffs They also examine the misses of 2025, including Siri and Alexa still being terrible, Apple’s underwhelming AI offerings, and the gap between promised AI agents and what actually exists today. Finally, Holly closes with a reminder that tech CEOs shape narratives as much as technology itself and that 2025 has been a year of learning to question the hype.

Episode 8: Inside the AI Agent Revolution

In this episode, Holly and Ewan explore one of the most hyped (yet deeply misunderstood) topics in AI today: AI agents. Holly opens with the big question: What actually is an AI agent? Ewan explains why definitions vary wildly, but broadly defines an AI agent as any system that can operate independently on your behalf to complete tasks. That could be a coaching assistant, a financial helper, or even a household or education agent. Ewan shares real-world stories, such as trying to buy a dishwasher using ChatGPT Agent Mode [https://openai.com/index/introducing-chatgpt-agent/]... Only to find that Amazon actively blocks agent-based access. When he switched to AO.com [https://ao.com/], the agent succeeded instantly - a perfect illustration of today’s fragmented ecosystem. He also discusses experimenting with agents to manage LinkedIn connection acceptance, with mixed results, highlighting how even simple point-solution tasks can quickly fall apart. The discussion then moves into the wider implications: * Why agents are transformational in theory, but fragile and unreliable today * How browser-based agents actually work using “computer use” screenshot loops * Why traditional RPA (Robotic Process Automation) remains far safer and more predictable * Early signs of agent-powered cyberattacks, referencing the first reported case of agentic hacking [https://www.anthropic.com/news/disrupting-AI-espionage] * The Carnegie Mellon “Agent Company [https://agent.company]” benchmark, which evaluates how well different agents perform real office tasks. With current leaderboards showing DeepSeek’s Matrix agent at ~43%, Google Gemini around 41%, and Claude Sonnet 4 around 33%. The conclusion? The vision is exciting, but today’s agents are nowhere near enterprise-ready. Expect rapid evolution, more experiments, and many more failures as this technology matures. If you've got feedback, we'd love to hear it. We reply to every single message! Find us at ⁠Working On It Podcast⁠ [https://www.workingonitpodcast.com/], or follow our ⁠LinkedIn Page⁠ [https://www.linkedin.com/company/wereworkingonit/]. Or talk to ⁠Holly⁠ [https://www.linkedin.com/in/digitalholly/] or ⁠Ewan⁠ [https://www.linkedin.com/in/ewanmacleod] on LinkedIn.

Episode 7: Are you buying a $499 home robot?

In this episode, Holly and Ewan debate the arrival of the $499-a-month X1 Neo humanoid robot [https://www.1x.tech/products/x1-neo]: A domestic helper that promises to tidy, load the dishwasher, and even put away groceries. Ewan’s enthusiasm meets Holly’s positive skepticism as they unpack the hype, the marketing fluff (“machine-washable knit suit,” anyone?), and the uneasy question of who’s really in control when a human operator can remotely pilot your household robot [https://www.1x.tech/blog/x1-neo-announcement]. From the reality of its four-hour battery life [https://techcrunch.com/2024/09/26/1x-launches-x1-neo-humanoid-robot-for-home-use/] to the privacy risks of a wandering camera, they explore where robotics sits today ... and how far it still has to go. The discussion also veers into more practical territory: robot lawn-mowers like Husqvarna’s Automower [https://www.husqvarna.com/uk/robotic-lawn-mowers/], Heathrow’s autonomous cleaning robots [https://www.heathrow.com/company/newsroom/blog/meet-heathrows-cleaning-robots], Abu Dhabi’s self-driving TXAI taxis [https://txai.ai/], and why Holly’s waiting for version 5 before inviting a robot into her home. Key topics: * The rise (and limits) of home robotics * What the $499/month X1 Neo [https://www.1x.tech/products/x1-neo] really does * Safety, privacy, and control concerns * Everyday robots that already work — vacuums, mowers, airport cleaners * The human cost and value of automation If you've got feedback, we'd love to hear it. We reply to every single message! Find us at ⁠Working On It Podcast⁠ [https://www.workingonitpodcast.com/], or follow our ⁠LinkedIn Page⁠ [https://www.linkedin.com/company/wereworkingonit/]. Or talk to ⁠Holly⁠ [https://www.linkedin.com/in/digitalholly/] or ⁠Ewan⁠ [https://www.linkedin.com/in/ewanmacleod] on LinkedIn.

Episode 6: How We Actually Use AI Every Day

In this episode of We’re Working On It, Holly and Ewan get personal about their real-world use of AI: Not the hype, but how these tools actually fit into their daily routines. Ewan breaks down his $300-a-month AI stack, from ChatGPT Plus and ChatGPT Pro [https://chat.openai.com] with ChatGPT Pulse [https://openai.com/pulse] to Anthropic’s Claude [https://claude.ai] and his WhatsApp-accessible assistant Martin [https://trymartin.com/]. He explains how each shapes his productivity, workflow, and even how he trains others to use agents effectively. Holly shares her own take, revealing how she uses Google Gemini [https://gemini.google.com] to summarise emails, ChatGPT for writing support and contract reviews, and why she still prefers to read over watch. Together, they discuss which tools have faded (farewell, Pi AI [https://heypi.com]), which have stuck (hello, Canva [https://www.canva.com]), and how AI has quietly become the new operating system for both work and home life. They also explore the economics of paying for AI subscriptions, how they use transcription tools like Otter.ai [https://otter.ai], and why AI accessibility — from boardrooms to gardeners — matters. Key topics: * Real productivity gains from AI assistants * Comparing ChatGPT Pulse [https://openai.com/pulse], Claude Code [https://claude.ai], Martin [https://trymartin.com], Gemini [https://gemini.google.com], and Canva’s Magic Studio [https://www.canva.com/ai/] * How AI tools are replacing older “point solutions” * The ethics and economics of AI subscription culture * Using AI for accessibility and empowerment If you've got feedback, we'd love to hear it! We reply to every single message! Find us at ⁠Working On It Podcast⁠ [https://www.workingonitpodcast.com/], or follow our ⁠LinkedIn Page⁠ [https://www.linkedin.com/company/wereworkingonit/]. Or talk to ⁠Holly⁠ [https://www.linkedin.com/in/digitalholly/] or ⁠Ewan⁠ [https://www.linkedin.com/in/ewanmacleod] on LinkedIn.