Beginners in AI

Good morning and thank you for joining us again!

Welcome to this daily edition of Beginners in AI, where we explore the latest trends, tools, and news in the world of AI and the tech that surrounds it. Like all editions, this is human curated, and published with the intention of making AI news and technology more accessible to everyone.

THE FRONT PAGE

Amazon’s House of David Used Over 350 AI-generated Shots

TLDR: Amazon's "House of David" embedded over 350 AI-generated shots into season two—and the show didn't collapse. Nor did the director when pushed about it.

The Story:

Amazon's biblical drama "House of David" used 350 AI-generated shots in season two—roughly four times what season one had. Creator Mark Boal didn't apologize for it. He called it production evolution, not a creative workaround. The show released. People watched it. Critics judged the actual storytelling. Nothing broke.

That's notable because it's the first major streaming series to openly announce AI at that scale and move on. Other AI uses in film got the quiet treatment: deepfaking Harrison Ford for "Indiana Jones and the Dial of Destiny," restoring a John Lennon vocal for a new Beatles track. Studios buried the methodology. Boal didn't. His message was clear—this is how you make television now.

The practical outcome shows up immediately in production work. AI isn't taking over cinematography or acting. It's building environments, extending backgrounds, solving visual problems that used to demand expensive location shoots, miniature teams, or months of post-production work. Timelines compress. Certain jobs disappear. Audiences watch the finished product and don't notice what built it.

Its Significance:

To understand where AI fits, you need the full timeline. There are really three types of effects work: practical effects (real explosions, stunts, animatronics built on set), CGI or digital effects (computer-generated imagery created in post-production), and now AI-generated content (algorithmic creation of environments, backgrounds, and visual extensions). Each wave replaced the previous one—and each time, entire job categories were disrupted.

But here's what's different about AI right now: it's not arriving as a separate tool. It's being woven directly into the software filmmakers already use every day. Editing suites. Color grading systems. Compositing software. Asset libraries. Within five years, you won't think about using "AI"—you'll just be using your editing software, which will have AI baked into it as a standard feature. It'll suggest transitions. It'll auto-generate missing frames. It'll fill in backgrounds. It'll be as invisible as the autocorrect function on your phone.

That's the real transition. Not "here comes AI, watch out." But "AI is becoming infrastructure." It stops being a choice about whether to use it and starts being an inevitability of the tools themselves.

Technological shifts in film used to take decades to absorb. Green screen didn't happen all at once. When it did arrive in the 1990s and 2000s, it gutted entire departments. Model makers who ran massive shops at Industrial Light & Magic watched their work evaporate as CGI solved what used to require physical miniatures. Set decoration crews and craftspeople who built physical environments? Those jobs went offshore to Canada, India, tax-advantaged countries. The labor market rewired itself. Filmmaking kept going. Audiences never knew the difference because the visuals got sharper.

Here's the thing though: audiences do care about quality when it matters. Background filler? Nobody notices if it's practical, CGI, or AI-generated. A sweeping landscape behind dialogue, an empty corridor, a crowd in the distance—viewers don't inspect the methodology. But throw that same AI or pure CGI into a major action sequence, a fight choreography, a hero moment where realism anchors the entire scene? Audiences absolutely notice. If it doesn't look convincing, they disconnect. The uncanny valley effect kicks in. People check out.

This distinction matters because it shows where AI actually has traction. It's not threatening the high-stakes, high-visibility work. It's automating the invisible stuff. Environment extensions. Background plates. Visual scaffolding that makes the real performances land harder. The work that used to require expensive location scouts, miniature teams, or post-production artists grinding for weeks.

Here's the counterintuitive part though: as those costs plummet, something else expands. Access. A teenager in Lagos, a film student in rural Poland, an aspiring director working a day job in Des Moines—they suddenly can produce visuals that looked impossible five years ago without a studio budget and a crew of dozens. They can't afford locations. They can't build sets. They can't hire a team. But they can use AI to generate environments, extend their limited locations, solve visual problems that used to be dealbreakers. This is the YouTube effect, or the podcast effect, if you will, being applied to visual storytelling.

That's a net positive nobody talks about. The same technology that compresses labor in Hollywood also democratizes filmmaking for millions of people trying to tell stories that matter to them. More voices getting heard. More perspectives getting made. More narratives escaping the gatekeeping bottleneck of expensive production.

AI's doing the same labor compression that CGI did. Faster. And filmmakers see it coming.

Denis Villeneuve—whose sci-fi films like Blade Runner 2049, Arrival, and Dune: Part Two basically redefined what big-screen cinema could look like—isn't actually worried about AI. In a Time magazine interview, he said he's more concerned about "the fact that we behave like algorithms, as filmmakers. We're in a very conservative time; creativity is restricted. Everything's about Wall Street. What will save cinema is freedom and taking risks."

He's onto something. The real problem isn't the tool. It's that studios already think like algorithms. They chase data. They avoid risk. They greenlight sequels over original stories. AI just gives them an excuse to keep doing what they were already doing anyway.

Christopher Nolan sees the danger differently. He doesn't hate AI—he views it as a powerful instrument. But he's explicit about one thing: "The person who wields it still has to maintain responsibility for wielding that tool. If we accord AI the status of a human being, the way at some point legally we did with corporations, then yes, we're going to have huge problems." His worry? Companies will use AI as cover. They'll blame the algorithm when things go wrong. They'll stop taking responsibility.

Both directors are describing the same risk from different angles. The technology's not the problem. Human choice is.

For "House of David," audiences demonstrated something counterintuitive: they care about story, performance, and direction. They don't care whether a background was painted, built practically, rendered in CGI, or generated by AI. Boal's show succeeds or failed based on its narrative and craft—not on the production pipeline. The 350 AI shots were invisible because they weren't the story.

TRENDING

Fei-Fei Li: Spatial Intelligence is AI's Next Frontier — AI pioneer Fei-Fei Li argues that the next major breakthrough in AI will be "spatial intelligence"—the ability to understand and interact with 3D space like humans do. Current AI can write and analyze text but struggles with basic tasks like navigating rooms or predicting physics.

AI Models Struggle with Basic CAPTCHA Tests — Researchers tested Claude, Gemini, and GPT-5 on solving Google CAPTCHAs and found that even the best model (Claude) only succeeded 60% of the time. GPT-5 performed worst at 28% because it spent too long thinking and timed out.

Firefox Cuts Browser Fingerprinting in Half — Mozilla's new Firefox 145 update blocks more ways websites track you by your device setup, cutting the number of users who can be uniquely identified from 35% to 20%. The protections work by standardizing how Firefox reports information like screen size and processor details.

AI Shows Same Thinking Mistakes as Humans — Researchers found that AI models like GPT-4 and Gemini make the same cognitive errors as human doctors when giving medical advice. For example, AI recommended surgery more often when survival rates were presented instead of death rates, showing the same "framing effect" bias that affects people.

WHERE WE STAND (Based on ☝️)

✅ AI Can Now: Beat human guesses at predicting what medical treatment an unconscious patient would want based on their background and values.

❌ Still Can't: Avoid making biased decisions that human doctors make, even when the mistakes are well-known.

✅ AI Can Now: Generate professional-looking fact-check articles complete with sources and citations that rank high in search results.

❌ Still Can't: Actually verify if the sources support the claims being made, leading to confident-sounding articles that spread false information.

✅ AI Can Now: Solve basic CAPTCHA tests designed to block bots, with success rates up to 60% on image recognition challenges.

❌ Still Can't: Handle time pressure effectively—models that think too long or check their work repeatedly end up timing out and failing tasks.

QUICK TAKES

Boba AI Labs Introduces Styles in Boba Anime

The story: Boba AI Labs released a new feature called Styles that lets anyone create anime videos in different visual styles like Shonen, Hand Drawn, American Comics, MSPaint, 90s Shojo, and more. Instead of just one anime look, creators can now choose from multiple artistic styles when generating their animations.

Your takeaway: This gives artists and creators way more creative control over how their AI-generated anime videos look, making the tool useful for different genres and storytelling styles instead of just one default look.

AI-Generated "Fact Checks" Spread False Information

The story: A website called Factually uses AI to automatically create "fact check" articles by searching the internet and asking language models to write summaries. Researcher Kyle Kingsbury found that Factually published false fact checks claiming a Chicago pastor wasn't shot by ICE agents with pepper balls in October, even though the incident was widely reported by major news outlets including Fox News and CNN.

Your takeaway: AI-generated fact checks can look professional and trustworthy while spreading completely wrong information, making it harder to know what's true online.

Uber, Lyft, and DoorDash Bet Big on Self-Driving Cars

The story: Uber, Lyft, and DoorDash each announced major investments in autonomous vehicles during recent earnings calls. DoorDash is spending several hundred million dollars more than planned on its robot delivery technology, including Dot, a stroller-sized robot that navigates sidewalks. Lyft is building a $10-15 million charging depot in Nashville for self-driving cars. Uber is currently losing money on driverless cars but plans to expand them as a key part of its business.

Your takeaway: These three major ride-hailing and delivery companies are committing huge amounts of money to autonomous vehicles because they see it as essential to their future, even though the technology isn't fully ready and won't completely replace human workers anytime soon.

CoreWeave Revenue Tops Estimates in Sign of Surging AI Demand

The story: AI cloud provider CoreWeave reported third-quarter revenue of $1.36 billion, beating Wall Street's prediction of $1.29 billion. The company's backlog of future contracts jumped to $55.6 billion, nearly double the $30.1 billion from last quarter. CoreWeave rents out powerful computer chips to companies like Meta and OpenAI who need them to run AI systems.

Your takeaway: The huge jump in future orders shows that demand for AI computing power is still growing fast, even though building enough data centers to meet that demand remains the biggest challenge.

AI Technology Race is New 'Cold War' Between US and China

The story: A new report from the Center for Security Policy warns that China could become the world's leader in AI by 2030, which would make the US a "second-rate power." The report says AI dominance matters because whoever controls AI could set worldwide technology standards, develop advanced military weapons, and potentially control how people access information.

Your takeaway: The next five years will decide which country leads in AI, and that decision will shape everything from military power to whose values get built into the technology billions of people use.

Netflix Releases Guidelines for Using AI in Content Production

The story: Netflix published official rules for when filmmakers can and can't use AI tools while making shows and movies. The guidelines allow AI for temporary work like planning scenes but require written approval before using AI to create anything that appears on screen, changes an actor's performance, or uses someone's likeness. Any AI use must be reported to Netflix contacts.

Your takeaway: Netflix is trying to embrace AI as a helpful tool while protecting actors, respecting copyrights, and making sure audiences can trust what they see on screen.

Apple Withdrawing Advanced Data Protection from UK

The story: Apple is removing its Advanced Data Protection feature from the UK because of government demands through the Investigatory Powers Act. Users who already turned on this privacy feature will have to manually turn it off or lose access to their iCloud accounts. The feature encrypted user data so even Apple couldn't read it.

Your takeaway: UK users are losing a major privacy protection because the government wants the ability to access encrypted data, forcing people to choose between privacy and using cloud storage.

RECOMMENDED LISTENING/READING

Ruins of the Earth by Christopher Hopper and J.N. Chaney is a masterclass in world-building and storytelling that'll hook you if you also enjoy high-action science fiction.

The AI character's wit absolutely steals the show—its personality is so compelling it makes you wonder if our future AI companions will have half this much charm. The audiobook narration by R.C. Bray is exceptional; this is definitely a listen rather than a read. I’m currently on book 4.

TOOLS ON OUR RADAR

• 🔨 Boba[Freemium]: Turn text descriptions into anime videos with voice acting, sound effects, and lip syncing in multiple anime styles.
  

• 🔨 Freepik AI Video Generator[Freemium]: Create professional videos from text or images using multiple AI models like Google Veo and Runway Gen 3, all in one platform.
  

• 📐 System76 Linux Laptops[Free OS only: Premium laptops built for developers with open-source Pop!_OS, upgradeable components, and open firmware for full control.
  

• 🔧 Privacy.com [Freemium]: Generate unique virtual card numbers for every online purchase to protect your real card info and block unwanted charges.
  

• 🪛 Notocat[Freemium]: Build a newsletter directly from your Notion workspace—write in Notion, upload to Notocat, and send to subscribers.

TRY THIS PROMPT (copy and paste into ChatGPT, Grok, Perplexity, Gemini)

I want to build a side hustle using AI tools to create income diversification. Create a 90-day action plan:

**My Situation:**
- Main job: [YOUR JOB TITLE]
- Industry: [YOUR INDUSTRY]
- Time available per week: [HOURS]
- Current skills: [e.g., writing, data analysis]
- Starting budget: [e.g., $0-500]

**Plan with these sections:**

1. **Three Side Hustle Ideas** (ranked by AI-leverage)
   - What you'd sell
   - How AI reduces time by 50%+
   - Monthly income potential (months 1-3, month 6+)
   - Specific AI tools needed
   - Why this interests you

2. **30-Day Launch Plan** (pick your top idea)
   - Week 1: Validate idea (find 5 potential customers)
   - Week 2-3: Build first deliverable using AI
   - Week 4: Launch and land first 3 customers
   - Pricing strategy

3. **AI Tools Workflow**
   - Tool #1: What it does
   - Tool #2: Time savings
   - Tool #3: Why essential
   - Show "Before (2 hours) → After (20 minutes)"

4. **Revenue Milestones**
   - Month 1: $[target]
   - Month 3: $[target]
   - Month 6: $[target]
   - What success means for you

5. **Risk Management**
   - Biggest obstacle
   - When to pivot
   - Exit strategy

6. **Time Boundaries**
   - Max hours per week
   - Work schedule
   - Reassessment frequency

Focus on the actual first step I'd take tomorrow—not generic advice.

What this does: Converts the week's AI job displacement reality (76,440 positions eliminated in 2025, entry-level jobs under pressure, but 170 million new roles by 2030) into a concrete anti-fragility strategy. Builds income diversification using AI tools that reduce time investment, perfect for beginners with limited availability. Might help you sleep better at night.

FROM THE WEB

This video flips the usual problem on its head. The physics is so convincing that people don't believe it was made with AI.

QUICK GUIDE Create and Edit in MidJourney Using Only Your Voice

Thank you for reading. We’re all beginners in something. With that in mind, your questions and feedback are always welcome and I read every single email!

-James

By the way, this is the link if you liked the content and want to share with a friend.