In partnership with

The skinny on what's happening in AI - straight from the previous live session:

Highlights

🤯 Fast Fashion Site Uses Luigi Mangione's Face for AI-Generated Modelling

Discussed at 09:38

Shein launched an investigation after Luigi Mangione's likeness was discovered modelling a £7.50 shirt on their website. The image showed a 99.9% facial similarity match according to BBC's facial recognition analysis, with telltale AI artifacts including dodgy hand rendering and blob-like anomalies. The company blamed a "third party vendor" and removed the image immediately.

Kyle's take: This is the most 2025-ass headline I've ever seen - AI generation, CEO killer, and fast fashion all colliding in one bizarre story. It’s like a story from The Onion but nope it’s just 2025!

The "third party vendor" excuse is getting old - everyone's trying to pass the buck when their AI content goes sideways. Honestly this is exactly why China's new watermarking laws make sense - traceability.

Source: BBC News coverage

💰 Free Perplexity Pro for a Year Through PayPal

Discussed at 16:50

This is not sponsored. Just cool!

Perplexity is offering 12 months of Pro access completely free when you connect a PayPal account. The Pro version gives you access to GPT-5, Claude 4.1 Opus, Gemini, and Grok - basically all the top models through one subscription instead of paying for multiple services. And it’s free for 12 month so…pretty good!

🚺 Women 20% Less Likely to Use AI Than Men

Discussed at 28:00

Analysis of 18 studies covering 140,000 people worldwide shows women are 22% less likely to use generative AI than men. Only 31.2% of Claude users and 42% of ChatGPT users are women. The gap exists across nearly all regions, sectors, and occupations, with women reporting lower awareness, less confidence with prompting, and being less persistent when outputs aren't satisfactory.

Kyle's take: This is concerning because it's not just about current usage - it's about the economic gap that's going to open up. If AI genuinely does boost productivity by 10x or 100x as everyone claims, then we're looking at massive income inequality between those who use it and those who don't. The Danish study mentioned shows women wanted more training but weren't getting it, while men just dove in and learned through mistakes. The fix seems to be better training programmes, but this isn't my space to preach - we need brilliant female educators leading this conversation to close the gap before it widens further.

Source: eWeek

Member Question from Mind & Glory: "I know there's a level of randomness with LLMs, but how can I stop getting different answers to the same prompt?"

Kyle's response: You can't, and that's both the wonderful and terrible thing about AI! The randomness (called stochastic behaviour) is what gives AI its creativity and human-like feel. If it was purely deterministic like traditional code, it would feel robotic. If you absolutely need identical outputs, specify extreme detail in your prompts and use JSON formatting for structured data. But honestly, if you need exact same input-to-output every time, don't use AI - use programming languages like Python instead.

Discussed at 13:53

Want the full unfiltered discussion? Join me tomorrow for the daily AI news live stream where we dig into the stories and you can ask questions directly.

Streaming on YouTube (with screen share) and TikTok (follow and turn on notifications for Live Notification).

Audio Podcast on iTunes and Spotify.

AI with Kyle - Tools, Resources and Products: Daily Show | Build an AI Business Challenge | AI Playbooks | 10 Week AI Summer Camp | BATON AI Business Series | AI App Building | AI Trainer Business Planner | What AI Business Should You Start? | AI Authority Accelerator | AI Automation Accelerator | 🏁 AI Workshop Kit