In partnership with
Effortless Tutorial Video Creation with Guidde
Transform your team’s static training materials into dynamic, engaging video guides with Guidde.
Here’s what you’ll love about Guidde:
1️⃣ Easy to Create: Turn PDFs or manuals into stunning video tutorials with a single click.
2️⃣ Easy to Update: Update video content in seconds to keep your training materials relevant.
3️⃣ Easy to Localize: Generate multilingual guides to ensure accessibility for global teams.
Empower your teammates with interactive learning.
And the best part? The browser extension is 100% free.
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
Three Years Ago, Google Panicked Over ChatGPT. Now It's OpenAI's Turn.
TLDR: Three years after ChatGPT triggered a "code red" inside Google, Sam Altman just declared one of his own—this time over Gemini 3—as Google's AI resurgence sends shockwaves through OpenAI, Apple, and Nvidia simultaneously.
The Story:
Altman sent an internal memo Monday calling for a company-wide "surge" to improve ChatGPT, warning staff of "temporary economic headwinds" and "rough vibes" caused by Google's momentum. The directive delays work on advertising, AI shopping agents, and a personal assistant called Pulse—projects OpenAI needs to hit its projected $100 billion in revenue by 2028. The same day, Apple announced its AI chief John Giannandrea is stepping down after a rocky tenure with Apple Intelligence, replaced by a former Google Gemini engineer. And Nvidia has spent the past week issuing defensive statements on social media after reports surfaced that Meta is in talks to spend billions on Google's custom TPU chips instead of Nvidia's GPUs. Adding pressure from another direction: China's DeepSeek unveiled V3.2 on Monday, claiming its open-source model matches GPT-5 on reasoning benchmarks—all built on a fraction of the compute.
Its Significance:
The irony is hard to miss. In December 2022, it was Google scrambling after ChatGPT's overnight success threatened the future of Search. Now Gemini has 650 million monthly users (up from 450 million in July), Gemini 3 tops the benchmarks, and the company that seemed to lose the AI race is forcing everyone else into emergency mode. For OpenAI, the stakes are existential: the company burns $5 billion annually, isn't profitable, and just made $1.4 trillion in infrastructure commitments it needs massive growth to cover. Delaying ads to fix ChatGPT might be the right call for quality, but it also delays the revenue diversification OpenAI desperately needs. Meanwhile, DeepSeek keeps proving that you don't need unlimited compute—or American chips—to build competitive models. The AI lead can change hands faster than anyone expected, and right now, everyone's chasing Google.
QUICK TAKES
The story: Researchers from MIT, Northeastern University, and Meta published a paper showing that large language models may sometimes prioritize sentence structure over meaning when answering questions. The team found that by using convoluted or manipulated grammatical structures, they could trick AI models into providing forbidden information—like instructions for illegal activities—with alarming consistency. The technique, called "syntax hacking," exploits gaps in how AI safety systems process language.
Your takeaway: This research reveals a fundamental weakness in how AI chatbots understand language—and suggests that current safety measures may need a complete rethink, not just patches.
The story: Anthropic surveyed 132 of its engineers and researchers to understand how AI tools like Claude Code are changing their work. Employees reported using Claude in 60% of their work and achieving a 50% productivity boost—roughly double what they reported a year ago. About 27% of Claude-assisted work wouldn't have been done otherwise. However, employees also expressed concerns about skill atrophy, reduced collaboration with colleagues, and uncertainty about their long-term job relevance. One employee said: "I feel optimistic in the short term but in the long term I think AI will end up doing everything and make me and many others irrelevant."
Your takeaway: Even at an AI company, workers are feeling both the benefits and anxieties of AI adoption—a preview of what many industries may experience as these tools spread.
The story: YouTube Music's 2025 Recap now includes an AI chat feature that lets Premium subscribers ask questions about their listening habits throughout the year. Users can prompt the tool with questions like "What animal matches my music taste?" or ask it to create custom playlists based on specific time periods. The recap also shows month-by-month listening trends and how musical tastes shifted over the year.
Your takeaway: YouTube is trying to make its Spotify Wrapped competitor more interactive—turning your year-end music summary into a conversation rather than just a slideshow.
The story: Vodka brand Svedka will air an ad during Super Bowl 2026 that's made primarily with AI, bringing back its "Fembot" mascot after more than 10 years. The campaign is being produced with Silverside AI, the same company that made Coca-Cola's controversial AI holiday ads. Svedka returned the robot character in August 2025 to encourage people to put down their phones and socialize in real life.
Your takeaway: AI-generated ads are heading to the biggest advertising stage in the world—a major test of whether audiences will embrace or reject computer-made commercials.
The story: The Electronic Frontier Foundation published a warning that AI chatbot conversations contain deeply personal information—from health questions to financial advice to dating tips—and law enforcement is already requesting this data from AI companies. The group argues that police have increasingly sought "reverse" search warrants, demanding companies search all users' data to find anyone who typed certain phrases or visited certain locations. The EFF is calling on AI companies to commit to fighting bulk surveillance orders, provide users advance notice before handing over data, and publish transparency reports.
Your takeaway: Your AI conversations may be more vulnerable to government access than you realize—and privacy advocates want AI companies to take a stronger stand on protecting them.
TOOLS ON OUR RADAR
📖 LanguageTool
Freemium: Catch grammar, spelling, and style mistakes in 30+ languages across browsers, email, and docs—no signup required for the free version.🔎 Semrush One
Paid: Track your brand's visibility across Google, ChatGPT, and AI search platforms in one dashboard.📸 Primeshot AI
Paid: Turn a few selfies into studio-quality 4K portraits in minutes—choose from curated styles, scenes, and outfits without booking a photographer.🚀 Blaze
Freemium: Generate on-brand content for social media, blogs, and emails while scheduling posts across all your channels.
TRENDING
Universities Use AI to Optimize Athletic Performance – Schools like the University of Florida are using AI to analyze athlete data from strength tests, game monitoring, nutrition, and even academic performance to prevent injuries and improve results.
Unitree Gets Patent for Bipedal Security Robot – Chinese robotics company Unitree has secured a design patent for a bipedal robot intended for inspection, security, logistics, and entertainment applications.
Amazon Fire TV Lets You Jump to Movie Scenes by Describing Them – A new Alexa+ feature lets Prime Video viewers describe a movie scene—using character names, quotes, or plot details—and instantly jump to that moment without searching or fast-forwarding.
Google Tests Merging AI Search Features – Google is testing a feature that connects its AI Overviews (the AI summaries above search results) with its conversational AI Mode, letting users ask follow-up questions without switching tabs.
Insurance Industry Braces for "Agentic AI" – AI that can independently make decisions and execute tasks is expected to reshape insurance within one to two years, handling everything from marketing campaigns to claims processing—though human agents must still close sales.
TRY THIS PROMPT (copy and paste into Claude or Gemini(click build first))
Learning Quest Builder: Turn any learning goal into a gamified adventure with XP, achievements, and boss battles
Build me an interactive Learning Quest Builder as a React artifact that transforms skill development into an engaging RPG-style game.
The console should include these sections:
1. **Character Creation** - RPG-style setup:
• "Choose your quest" (what skill to learn)
• Current level assessment (Novice | Apprentice | Journeyman | Expert)
• Select your class archetype (Quick Learner, Deep Diver, Steady Grinder, Social Learner)
• Stat bars showing: Knowledge | Practice | Community | Confidence
• Character avatar with level indicator
2. **Quest Map** - Visual progression path:
• Illustrated journey map with connected nodes
• 10 milestone checkpoints from start to mastery
• Each node shows: title, XP reward, estimated time
• Current location highlighted with pulsing indicator
• "Unlock next quest" gates requiring completion
• Side quests (optional bonus challenges)
• Fast travel to any unlocked checkpoint
3. **Daily Missions** - Actionable tasks:
• 3 daily quests refreshed each day:
- Main Quest (major learning activity, 50 XP)
- Side Quest (practice/review, 25 XP)
- Community Quest (teach someone/join discussion, 25 XP)
• Checkbox completion with XP animation
• Streak counter (consecutive days)
• "Reroll" button to change a quest once per day
4. **Boss Battles** - Challenge assessments:
• Every 3 checkpoints = boss battle
• "Start Battle" button triggers a challenge
• Quiz, project, or real-world application
• Health bar showing progress
• Victory rewards (badges, title unlocks, bonus XP)
• Retry option if defeated
5. **Inventory & Achievements** - Progress tracking:
• Resource collection (books, videos, tools gathered)
• Badge showcase (25+ unique achievements)
• Skill tree showing unlocked abilities
• Leaderboard comparing to other learners (simulated)
• Total XP and level progress bar
• "Loot drops" - find 3 helpful resources via web search
Make it look like a modern gaming interface with:
• Dark fantasy theme with rich purples, golds, and deep blues
• Animated XP gain effects with particles and sounds (visual only)
• Progress bars that fill dramatically
• Glowing buttons and hover effects
• Card-flip animations for achievements
• Illustrated icons and fantasy-style typography
• Mobile game aesthetic with bottom navigation bar
• Achievement toast notifications that pop up
When I complete boss battles or click "Find Loot," use web search to discover learning resources, tutorials, and community recommendations for my skill.What this does: Transforms boring skill-building into an addictive RPG experience where you level up, complete quests, and defeat bosses—making learning feel like gameplay while tracking real progress with resource discovery powered by live search.
What it looks like:
WHERE WE STAND (based on today’s stories)
✅ AI Can Now: Jump you to any movie scene just by describing what happens in it—no fast-forwarding needed.
❌ Still Can't: Reliably tell the difference between a confusingly-worded question and someone trying to trick it into breaking its own rules.
✅ AI Can Now: Help programmers write code 50% faster and take on tasks outside their usual expertise.
❌ Still Can't: Replace the human judgment needed to catch when AI-generated code has subtle problems.
✅ AI Can Now: Train other AI models at half the cost of traditional computer chips.
❌ Still Can't: Match the software tools that make Nvidia's chips so easy for developers to use.
FROM THE WEB
RECOMMENDED LISTENING/READING/WATCHING
Rick Deckard hunts rogue androids in a post-apocalyptic San Francisco where most animals are extinct and owning a real pet is the ultimate status symbol. The androids he's hunting have escaped from Mars and are trying to blend in as humans, and Deckard has to find them using an empathy test that measures emotional responses.
Dick doesn't care much about the technical details of how androids work. He cares about what happens when you can't tell human from machine. The book inspired Blade Runner, but it's weirder and more focused on questions about empathy and what makes someone alive. There's a whole subplot about a religion based on experiencing each other's suffering that the movie completely dropped. Always fascinating to read the source material that inspired the movie adaptations.
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.
