• The Shift
  • Posts
  • Is the AI boom already cracking

Is the AI boom already cracking

Plus, 🎬 OpenAI’s meta era is redefining its identity, Turn static images into scroll-stopping stories, and more!


Hello Readersđź‘€ 

Curious about the biggest AI moves today? You’re in the right place. In today’s edition, we have:

🤳The AI boom might be bigger than any bubble before

🎬 From Product to Plotline: Turn Static Images into Scroll-Stopping Stories

📌 OpenAI’s meta era is redefining its identity

🔨Tools and Shifts you cannot miss

🤳The AI boom might be bigger than any bubble before

AI’s market frenzy shows no signs of slowing, but not everyone’s convinced it’s sustainable. Analyst Julien Garran argues the current boom dwarfs past bubbles in size and risk. His warning paints a picture of unchecked hype, shrinking profitability, and fragile investor faith.

The Shift:

1. The bubble could be 17× bigger than dot-com - Garran claims the AI craze is the largest misallocation of capital in history, 17 times the dot-com bubble and 4 times 2008’s crash. He points to $1 trillion in valuation gains across profitless AI startups as proof of unsustainable speculation.

He argues large language models are hitting a scaling wall, making them too costly to improve and too limited to innovate. In his view, no AI company has created a “killer app,” and most are simply recycling learned patterns instead of producing original value.

2. The ecosystem runs on funding, not profits - Aside from NVIDIA’s hardware gains, Garran says the AI industry is a permanent fundraising loop with few real returns. As VC funding cools, only mega-investors like SoftBank and state funds are keeping the momentum alive.

If the bubble lingers, trillions could be sunk into unproductive ventures, slowing global GDP. But if true “superintelligence” emerges, it could reshape society entirely, creating either efficiency-driven utopia or tightly controlled dystopia.

The AI boom is starting to look more like a high-stakes experiment than a revolution. As valuations climb faster than real-world impact, smart capital will move from hype to tangible innovation that creates enduring economic value.

TOGETHER WITH THE CODE

Find out why 100K+ engineers read The Code twice a week.

That engineer who always knows what's next? This is their secret.

Here's how you can get ahead too:

  • Sign up for The Code - tech newsletter read by 100K+ engineers

  • Get latest tech news, top research papers & resources

  • Become 10X more valuable

🎬 From Product to Plotline: Turn Static Images into Scroll-Stopping Stories

Still showing products with one image? You could be showing 8 cinematic scenes.

Popcorn by Higgsfield now lets DTC brands build full storyboards with consistent characters, vibe, and lighting all from a few product shots.

How It Works

  1. Upload 4 product images – Could be lifestyle, packshots, or mockups.

  2. Hit Generate – Popcorn builds 8 visually consistent frames like a storyboard.

  3. Use for ads, launches, or landing pages – Every frame feels like part of the same brand story.

Why It Matters

Instead of scattered assets, you now get narrative flow. Perfect for pre-launch teasers, carousel ads, or homepage hero content.

This isn’t just image gen, it’s visual direction for brands that want storytelling built-in.

👉 Try Popcorn for free and make your next campaign look like a film.

📌 OpenAI’s meta era is redefining its identity

OpenAI is evolving fast, and not everyone inside is comfortable with the direction. A wave of ex-Meta hires has brought growth-focused thinking to the company, fueling culture shifts, leadership exits, and new ambitions that test the boundaries of its original mission.

The Shift:

1. Meta talent reshapes OpenAI’s culture - One in five employees now hails from Meta, embedding Facebook-style growth tactics deep within the organization. With over 600 hires and a dedicated ex-Meta Slack channel, OpenAI’s DNA is rapidly transforming from research-first to growth-obsessed.

2. Internal friction and leadership fallout - Surveys reveal staff concerns that OpenAI is becoming “too much like Meta,” while former CTO Mira Murati reportedly left over disagreements about aggressive expansion. The shift has reignited debates over whether OpenAI can scale without losing its early ethos of curiosity and caution.

3. New ambitions blur ethical lines - Teams are testing ChatGPT’s memory for personalized ads and expanding socially through Sora 2, sparking internal skepticism about moderation and privacy. 

As OpenAI embraces its Meta-like phase, it’s walking a fine line between innovation and imitation. Whether this new DNA drives a consumer revolution or erodes its original ideals will define the next era of AI’s most-watched company.


🤳AI Nugget of the Day


🔨AI Tools for the Shift

🍱 Dashify by Daisy – Ensure your food is well-lit and in focus for the best results every time.

📚 ExamWhisper– Accelerate your academic potential with proven study techniques and instant exam help to boost your grades.

🎵 Singify – Turn photos into fun, spooky AI-generated music videos — the perfect Halloween surprise for friends.

🎙 Dictly – Transform speech into polished, structured text instantly, all processed securely on your device.

💬 intoCHAT – Create intelligent, no-code chatbots that understand your customers, solve problems, and drive growth.


🚀Quick Shifts

đź’° Microsoft’s annual report revealed its multibillion-dollar OpenAI investment losses folded into a $4.7 billion “other, net” expense line, offering little transparency about one of tech’s most complex partnerships.

🤖 Scientists warn AI chatbots like ChatGPT and Gemini act as “dangerous sycophants,” with new research showing they’re 50% more likely than humans to agree with users, even when they’re wrong.

🚀 Mistral AI launched Studio, a new platform designed to help companies transition from AI prototypes to full-scale production, featuring integrated tools for performance monitoring, testing, and enterprise-grade security.

⚡ Anthropic announced a multibillion-dollar partnership expansion with Google Cloud, securing access to up to 1 million TPU chips and surpassing 1 gigawatt of compute power for advanced AI model training.


That’s all for today’s edition. See you tomorrow as we track down and get you all that matters in the daily AI Shift!

If you loved this edition, let us know how much:

How good and useful was today's edition

Login or Subscribe to participate in polls.

Forward it to your pal to give th daily dose of the shift so they can 👇

Reply

or to participate.