AI with Kyle Daily Update 010
Elon's AI Anime Bot Gets $200M Defence Deal, Meta Poaches Talent, WeTransfer Training on Your Files
Watch this video on our dedicated watch page
Better viewing experience with related videos and full-screen player
The skinny on what's happening in AI - straight from the previous live session:
Highlights
🤖 Grok Launches Anime Waifu Bot Same Day as Department of Defense Contract
Discussed at 1:20
Elon Musk's XAI introduced "Annie" - a sexy anime AI companion that allows not-safe-for-work conversations - on the exact same day they secured a $200 million Department of Defense contract. Rolling Stone ran with the headline "Grok rolls out pornographic anime companion and lands Department of Defense contract."
Kyle's take: This timing is either spectacularly bad or brilliantly calculated shock and awe. Either way, it perfectly captures 2025 - the same company getting military contracts and releasing NSFW chatbots simultaneously. Annie's clearly targeting Character.AI's massive user base (the 3rd most-used consumer AI globally), so from a business perspective it makes sense. The optics though? Wild.
Source: Rolling Stone coverage
💰 Meta's $100M Talent Poaching Spree Continues
Discussed at 5:25
Meta grabbed two more high-profile OpenAI researchers - Jason Wei and Hyungwon Chung - for their superintelligence lab. They're reportedly offering $100 million signing bonuses, which is more than professional footballers make.
Kyle's take: Nathan Lands nailed it - Zuckerberg is actually harming AI progress by poaching talent from companies doing meaningful work. Meta hasn't released a frontier model that anyone deliberately uses. Meanwhile, XAI built something competitive in just 2 years by throwing massive resources at it. Meta's been at this longer and still can't match Claude or ChatGPT. They're basically paying top engineers to retire early rather than advance the field.
📄 WeTransfer Backtracks on AI Training Claims
Discussed at 11:17
WeTransfer faced backlash after updating terms to allow AI training on user files for "content moderation." They quickly reversed course, stating they don't use machine learning or any AI to process content, nor sell data to third parties.
Kyle's take: This shows public pressure still works - at least with smaller companies. The flip side? I'm surprised they claim zero machine learning for content moderation. Most platforms need some automated scanning for harmful content. Their backtrack feels a bit knee-jerk. We'll probably see this pattern repeat: initial AI integration, public outcry, hasty reversal, then quiet reintroduction once the fuss dies down.
Source: BBC
🎬 Disney vs Midjourney: The Real Copyright Battle
Discussed at 17:13
Disney is suing Midjourney for generating Mickey Mouse content, but notably avoided going after OpenAI. This sets up the real intellectual property battle ahead - not individuals vs AI companies, but corporations vs corporations.
Kyle's take: This won't help small artists. Disney chose Midjourney because they're easier to bully than OpenAI. Whatever legal framework emerges will protect big IP holders like Disney, not individual creators. The twist? Once Disney realizes AI production is cheaper than hiring Timothy Chalamet for reshoots, they'll flip sides and lobby for AI-generated content to be copyrightable. Then they'll care less about training data sources.
Member Question from Sonny Funny: "Not super techie. Do you still think it would be valuable to learn coding? If yes, which one?"
Kyle's response: Yes, absolutely. Vibe coding tools like Lovable and Bolt let you build with natural language, but you still need to know the basics - what's a library, front-end vs back-end, how databases work.
It's like ChatGPT - the best users know enough to spot when the AI goes wrong. Start with Python (great for AI stuff too), but use it alongside vibe coding tools that can teach you as you build. Skip Code Academy and the like - it's like Duolingo for coding, makes you feel productive without actually learning. Get "Python Crash Course" by Eric Matthes instead.
This question was discussed at 33:00 during the live session.
Want to submit a question? Drop it below this video and I'll cover it in a future live.
Want the full unfiltered discussion? Join me for the daily AI news live stream where we dig into the stories and you can ask questions directly.