🤖 ChatGPT Faces Its Toughest Competition Yet

Hello AI Enthusiast,

OpenAI just hit the panic button. Sam Altman sent an internal "code red" memo this week, pausing new features to focus on improving ChatGPT as Google and Anthropic catch up fast. The company also faced a security incident involving analytics provider Mixpanel that exposed some API user data. And MIT just released a tool that maps exactly which jobs AI can already replace. Let's break down what's happening.

The Big Picture 🔊

OpenAI Declares 'Code Red' as Rivals Close Gap

Sam Altman announced an internal "code red" to refocus OpenAI on improving ChatGPT's speed, reliability, and personalization. The move delays other products like AI agents and advertising as Google's Gemini 3 and Anthropic's Opus 4.5 outperform GPT-5 on benchmarks. Despite maintaining 800 million weekly users, OpenAI faces mounting competitive pressure.

OpenAI Reports Security Incident

OpenAI disclosed that Mixpanel, a web analytics provider, was breached on November 9. Limited API user data was exposed, including names, emails, approximate locations, and organization IDs. No ChatGPT users, passwords, API keys, chat data, or payment information were compromised. OpenAI terminated its Mixpanel contract.

MIT Study: AI Could Replace 11.7% of U.S. Jobs

MIT's new Iceberg Index simulates AI's impact on 151 million U.S. workers, finding current AI can already replace 11.7% of jobs—$1.2 trillion in wages across finance, healthcare, and professional services. The tool maps 32,000 skills across 923 occupations, revealing exposure extends beyond tech hubs to routine functions in HR, logistics, and administration.

Bits and Bobs 🗞️

• Runway Gen-4.5 new AI model generates high-definition videos and leads Google's leaderboard, challenging major companies with its advanced efficiency.

• Black Forest Labs' FLUX.2 advances AI image generation with an open-source latent space VAE, multi-reference conditioning, and improved text rendering.

• Mistral developed efficient open-weight language models, including new small models that run on a single GPU.

• Anthropic's Claude Agent SDK improves long-running AI coding tasks by using incremental progress tracking, JSON-based feature management, and automated testing to maintain context across sessions and enhance code quality.

• Claude's estimates suggest that broad adoption of current AI systems could boost US labor productivity by about 1.8% annually over ten years.

• Nvidia released Alpamayo-R1, a new AI model on GitHub designed to improve autonomous vehicle decision-making and support physical AI in robots.

• Booking.com developed a multilingual Generative AI agent using GPT-4 Mini to automatically suggest safe and accurate responses to guest inquiries.

• Google and OpenAI have imposed stricter limits on free AI tool usage, encouraging users to upgrade to paid plans due to high demand and heavy computing costs.

• The USPTO clarified that AI is considered a tool in invention creation, and only human inventors can be credited for patents under existing standards.

From Our Founder’s Channels 🤳

In his recent LinkedIn post, Gianluca reflects on ChatGPT’s 3rd-birthday statistics, highlighting the remarkable speed of global adoption and suggesting that the true impact of AI is still widely underestimated.

That's a wrap on our newsletter! Before you go, here’s a quick recap of our offerings:

• AI Academy Membership: Get 12 months of access to all our cohort-based programs, live webinars, on-demand courses, and tutorials.

• AI Agent Bootcamp: Accelerate processes and solve business problems by mastering prompts and building AI Agents, without coding.

• Practical Introduction to ChatGPT: A free course on using ChatGPT confidently, understanding its workings, and exploring its potential.

• Customized Corporate Training: Equip your team with the skills they need to unlock the potential of AI in your business.

Catch you next week! 👋