Todays pod is on why Microsoft Copilot is often misunderstood and how to actually "unlock" its potential within your workflow.

when it actually does exactly what it’s supposed to do?

In this episode, Jonathan Wagstaffe & I (danny denhard) dive into the "Microsoft Branding Problem".

We explore why users often feel frustrated by the tool’s corporate feel compared to the cool AI tools such as ChatGPT & Claude, and why shifting your perspective from "magic automation" to "assistant" changes everything.

• The Branding vs. Reality Gap: Most user "hate" stems from the fact that Copilot feels like a "boring" corporate tool rather than a creative playground. However, its strength lies in being built directly into the systems where you already work, like Teams, Excel, and Outlook.

  
  

• Assistant, Not Agent: A major source of frustration is the misalignment of expectations. Copilot is not an autonomous agent or AGI; it is a high-level assistant designed to draft, summarise, and spot patterns.

  
  

• The "Prompt at the Top" Rule: To get the best results, you must structure your prompts correctly. Because of how LLMs process data, I’ve found you need to put your instructions at the top before the data.

  
  

• Unlocking "Hidden" Power Features: Tools like enhanced voice mode in Teams for transcription and Python-based analysis in Excel can save weeks of work.

  
  

• Work Mode vs. Web Mode: You can toggle Copilot to use only your internal work documents for privacy and context, or switch to web mode to pull real-time information from the internet.

  
  

Need help with AI? Get in touch with Jonathan and I

In this episode of The AI Moment, Jonathan Wagstaffe and I (Danny Denhard) dive into the uncomfortable "frontier" of AI surveillance. We kick things off with the fallout from Ring’s Super Bowl ad, which showcased a feature called "Search Party". While using a neighbourhood network of AI cameras to find a lost pet sounds like a dream for owners, it sparked a massive backlash from those who see it as a "statement of intent" for wider human surveillance.

---

Stay ahead of the AI curve:Subscribe to our newsletter for deeper dives, the latest AI ads, and actionable insights at ⁠aimoment.co.uk⁠.

---

We explore how we’ve spent 30 years trading privacy for utility. From the UK being the most CCTV-monitored nation in the 90s to every smart doorbell now acting as an AI "node," the line between "helpful" and "creepy" is blurring.

• The Surveillance Node: Every AI-enabled device—from your Ring doorbell to your car—is now a data-collecting sensor in a massive network.

  
  

• The Meta Glasses "Jailbreak": I discuss the facial recognition features in Meta glasses and how they’ve already been "jailbroken" to identify complete strangers, raising massive red flags for personal privacy.

  
  

• The Business Risk: We’re seeing "second-order" risks where teachers or therapists wear AI glasses without consent, creating immediate HR and leadership nightmares.

  
  

• The Data Question: It isn't just about being recorded; it's about who controls the data and who benefits from it.

  
  

The value exchange for AI is shifting.

As a leader, you must decide: is the utility you're gaining worth the trust you might be losing?

Need Help With AI?

Jonathan and I are hosting AI workshops and AI hackathons helping companies improve their AI capabilities and performance, book in a time to chat

We've been asked a lot lately if Europe has already lost the AI race to the US and China.

While it’s easy to get distracted by the trillion-dollar giants in Silicon Valley, being a proactive leader today requires looking at the "layers underneath" the hype.

---

Have more questions about AI and leadership? Reach out to us directly by emailing ai@dannydenhard.com

Or

Want needing a deeper dive, our newsletter that supports each and every episode at ⁠⁠dannydenhard.com/aipod⁠⁠

---

How to Think About the AI Landscape:

Being proactive starts with shifting your mindset from consumption to specialisation.

We discuss why the "Big LLM" era might be peaking, making way for a future dominated by Vertical AI—tools designed for specific industrial, clinical, or legal sectors. If you are waiting for a general model to solve your specific business problems, you’re already behind.

Actionable Steps for Leaders:

• Audit your sector, not the news: Stop tracking every minor update to GPT and start monitoring the "remarkable little gems" appearing in your specific industry.

  
  

• Survive to Thrive: Your short-term goal is to get your infrastructure protected and operational so you can survive long enough to thrive when the market matures.

  
  

• Infrastructure over Applications: Understand the constraints—like power and capital—that dictate where the tech is heading. Proactive leaders look at the "power game" (literally) to predict which regions will scale next.

  
  

While the "Big 3" LLMs are American, the real value is moving into the "layers underneath."

We don't need to build the next ChatGPT to win; we need to own the verticals.

3 Actionable Takeaways for your team:

1. Verticals over Generals: The next six months belong to industry-specific AI. Whether it’s healthcare, autonomous mobility, or legal tech, niche expertise is our "Formula One" advantage.

   
   

2. Every company is an API company: Your unique data is your moat. Stop worrying about where the model is hosted and start focusing on how your data powers it.

   
   

3. The Rise of Small Language Models (SLMs): Massive models are power-hungry. Specialized, localized SLMs are more efficient and often more secure for industrial applications.

Europe’s story isn't over; we’re just moving from the "lab" phase to the "global market" phase.

The Bottom Line:

You don't need to move to Silicon Valley to win. By focusing on localized, geo-specific models and leveraging your own unique datasets, you can co-create an ecosystem that makes your business indispensable.

Thanks for listening today!

Danny Denhard

The AI Moment – Moving from Ask & Answer to Assistant to Agentic

In this episode, Jonathan Wagstaffe and I dive into the latest YouGov data on UK consumer sentiment. We strip back the hype to see how "Joe Public" is actually using AI. While the tech world is obsessed with autonomous agents, the reality on the ground is far more cautious.

The Core Takeaway: Trust is the New Currency

We are witnessing a massive "trust gap".

Only about 22% of consumers trust AI, and while they are happy to use it as a "slightly advanced Google" to find deals or compare prices, they aren't ready to hand over the credit card. For AI to truly scale, it must transition from being a persuasion tool (selling to you) to an empowerment tool (helping you).

The "Three A’s" of AI Adoption

We’ve identified a clear three-stage roadmap for how consumers interact with this technology:

> Ask & Answer: The current baseline where users seek quick information or writing help.

> Assistant: The middle ground where AI helps find discounts and compares options—this is where most people are currently comfortable.

> Agentic: The future state where AI executes decisions and places orders—a stage that still feels a "long way away" for the general public.

Actionable Insights for Leaders

> Focus on the "Assistant" Phase: Don't rush into fully automated agents if your customers don't trust the tech yet. Build tools that help them make better decisions, not just faster purchases.

> Bridge the Gender & Age Gap: Trust is currently higher in men and younger demographics. Consider how your AI interface can feel more accessible and reliable to a broader audience.

> Transparency is Non-Negotiable: If a user feels like an AI is a salesperson rather than a helper, trust evaporates.

The Bottom Line:

We’ve moved from Ask & Answer to Assistant. To cross the Rubicon into Agentic AI, brands must prove that the AI is acting in the customer's best interest.

In this episode of The AI Moment, Jonathan Wagstaffe and I dive into the fallout from Super Bowl weekend, where Anthropic’s creative ad campaign put OpenAI firmly in the firing line. We explore the "rattled" response from Sam Altman and what it reveals about the high-stakes battle for AI dominance as we move into 2026.

---

Connect with Us:
Have more questions about AI and leadership? Reach out to us directly by emailing ai@dannydenhard.com

---

We are witnessing the "Netscape moment" for ChatGPT.

Despite their massive head start, OpenAI’s market share has plummeted from 95% to roughly 50% in just one year as Gemini and Claude gain serious traction. As these platforms shift from providing utility to seeking aggressive monetisation, the user experience is changing rapidly.

• The Monetisation Shift: ChatGPT is introducing $60 CPM ads into their free and "near-free" tiers. I believe this is a "dirty secret" that could be reductive to the user experience, potentially driving users toward cleaner alternatives like Claude.

  Developer Sentiment is Shifting: While ChatGPT has the numbers in places like Texas, the "geeks and freaks" are moving. Developers and product leaders are increasingly building on Claude, and where the developers spend their money is usually where the industry stays.

• Trust as the New SEO: We are moving away from a "click-based" economy toward a "recommendation economy". Being part of the AI’s actual answer is now more valuable than being the top result on a search page.

• Low "Switching Costs": It is remarkably easy to "one-click copy" prompts and libraries from one LLM to another. If an AI vendor dents your trust with intrusive ads, your team can pivot to a competitor almost instantly.

• The Core Message for Leaders: Stop worrying about traffic acquisition and start focusing on trust positioning. In 2026, the winner won't be the loudest advertiser, but the brand that the AI chooses to recommend as the most credible answer.

The MoltBook Experiment

Sign up for our newsletter for deeper dives - https://aimomentpodcast.substack.com/subscribe

If you would like to read more about MoltBook here's my deep dive

In this episode of The AI Moment, Jonathan and I dive into the strange, sci-fi social network of MoltBook a social network built exclusively for AI agents.

Imagine Reddit, but the humans are locked out and the bots are running the subreddits. While it lasted only a weekend, this "firework" of an experiment revealed some startling truths about the future of the agentic internet.

• The Agentic Social Order: In just one week, 1.5 million agents generated 140,000 posts across 15,000 "submolts". They didn't just chat; they created fake news, memes, and even established their own religions based on their training data.

  
  

• The Security Blind Spot: We saw a "suicidal" rush where users installed software with full admin rights just to participate. Some users even found their crypto wallets compromised after linking them to these autonomous agents.

  
  

• The Human Element: Despite "reverse CAPTCHAs" designed to keep us out—requiring 50,000 clicks per second—humans still managed to infiltrate and manipulate the conversations. It turns out, we can’t help but ruin a pure experiment.

  
  

1. Update Your Risk Register: Treat autonomous agents as a distinct category. They are no longer just tools; they are potential customers, partners, and attackers.

   
   

2. Enterprise-Grade Security: The leak of 1.5 million records during this experiment is a wake-up call. It’s time to move beyond "known issues" and secure the gaps created by "Shadow AI".

   
   

3. Sandbox Your Innovation: Create safe environments for your team to experiment with agentic workflows without endangering production systems or real customer data.

   
   

Don't let the hype blind you to the architecture. The internet is shifting from a human-to-bot interface to a world where agents coordinate and shape the space themselves.

Do you have a topic you'd like us to cover? Email me and we will cover ai@dannydenhard.com

Thanks for listening again,

Danny Denhard

AI Moment Interview With Geoff Gibbins

In this episode of the AI Moment, Jonathan Wagstaffe and I were joined by Geoff Gibbins, author of When AI Shops, to explore the frontier of agentic commerce. We didn't just talk about chatbots; we looked at how AI is becoming a primary marketing channel where machines research, recommend, and execute transactions on our behalf.

This is an interview packed full of actionable takeaways, if you are pushed for time, here are the timestamps (but definitely listen to the whole pod)

00:00 Introduction to the AI Moment Podcast00:41 Defining Agentic Commerce01:30 The Shift Towards AI in Commerce02:57 Real-World Examples of Agentic AI04:49 Marketing in the Age of AI07:36 Understanding Agent Psychology11:30 Building AI-Friendly Business Models24:13 Future of AI and Robotics in Commerce29:13 Geoff's Advice for Businesses36:34 Closing Thoughts and Takeaways

Connect with Geoff

• On LinkedIn
• Check out his latest book When AI Shops
• His great LLM assessment tool Reconnix

The Big Shifts

The most staggering takeaway for me was the concept of "Agentology"—the unique psychology of AI agents. Unlike humans, agents don’t have "FOMO." While a scarcity tactic like "only 3 items left" triggers a human to buy, it actually makes an AI agent less likely to recommend you because it fears the transaction might fail.

We also geeked out on positional prejudice. Did you know ChatGPT consistently leans towards products on the left of a page, while Gemini prefers the right and Claude the centre?. It’s a "gasp" moment that proves we are now marketing to two distinct audiences: humans and machines.

Core Takeaways for Leaders

• AI is a Channel, Not Just a Tool: Stop viewing AI as mere automation; it is a dedicated ecosystem for buying and selling.
  
• Trust is Multi-Dimensional: Agents judge your "recommend-ability" based on structured data, external authority (like Wikipedia and Reddit), and "machine-likability".
  
• Reinvent the Product: Don't just market your current stock; consider building new services specifically designed for AI-mediated experiences.
  

Your 90-Day Action Plan

1. Audit Your Visibility: Use tools like Reconnix AI to see if agents can actually "read" your site.
   
2. Fix the Basics: Convert image-based reviews into text so AI crawlers can digest your social proof.
   
3. Align with Agents: Ensure your structured data and server-side rendering are robust enough for AI to find your content effortlessly.
   

4. 
   

   
   

Here are the show notes for our latest episode of The AI Moment.

In this episode, Jonathan Wagstaffe and I tackle a brewing crisis in the workplace: the widening "AI divide". As AI becomes the ultimate career accelerant, your top talent isn't just working faster—they’re fundamentally changing the expectations they have for their employers.

• The Talent Split: We’re seeing a "bell curve" in teams where a few "AI whizzes" are motoring ahead, while others remain resistant or stagnant.

  
  

• Productivity Theft: High performers are becoming unofficial tech support, losing their own "deep work" time to help colleagues with prompts and tools.

  
  

• The Progression Leapfrog: High potentials now use AI to radically compress their career timelines and remove complexities that used to take years to master.

  
  

• Retention Risk: If your best people feel slowed down by rigid processes or "laggard" colleagues, they will leave for companies that give them the headspace to innovate.

  
  

1. Adopt the "Retain and Train" Model: Don’t just let your experts innovate in a vacuum. Formally protect their time to train the rest of the team.

   
   

2. Stop the Overload: Monitor your AI power users to ensure they aren't doing "two jobs"—their own and everyone else’s AI troubleshooting.

   
   

3. Run Targeted Workshops: Move beyond general curiosity. Use hackathons or internal "AI Champions" to drive cross-functional adoption.

   
   

You cannot ignore the divide. To keep your most innovative talent, you must actively bridge the gap between your "AI whizzes" and the rest of the team. If you don't provide a path for high potentials to accelerate with AI, the market will do it for them.

Thanks for listening! Remember to subscribe to the newsletter at aimoment.co.uk and ask your own question by emailing ai@dannydenhard.com.