- The Recap AI
- Posts
- DeepMind's AI coding champion
DeepMind's AI coding champion
PLUS: China's complete Nvidia ban and a new AI model that predicts your future diseases
Good morning, AI enthusiast.
A version of Google's Gemini 2.5 has achieved a new milestone in a major programming competition. The model not only won a gold medal but also solved a complex problem that no human competitor could crack.
The achievement is significant because it shows the model's ability for abstract reasoning, not just mastering a specific game. With this leap, could we be on the verge of AI accelerating major scientific and engineering breakthroughs?
In today’s AI recap:
Google’s Gemini model solves impossible code
China’s ban on Nvidia AI chips
A new AI model predicts your future health
8 trending AI Tools
Google's AI Coding Champ

The Recap: Google DeepMind announced a version of its Gemini 2.5 model achieved a historic performance at the International Collegiate Programming Contest, winning a gold medal and outperforming most of the world's top human programmers.
Unpacked:
Gemini placed second overall among 139 teams, correctly solving 10 out of 12 complex algorithmic problems within the five-hour competition.
It was the only competitor to solve one complex problem involving multi-dimensional optimization, a puzzle that stumped every human team.
The achievement is notable because it used an enhanced version of a general-purpose model, demonstrating a profound leap in abstract reasoning rather than performance on a narrowly trained task.
Bottom line: This marks a significant step for AI, moving beyond pattern matching in games to tackling abstract, real-world problems. This advanced reasoning capability could soon help accelerate breakthroughs in fields like engineering and drug discovery.
You’re invited to the world’s largest email marketing conference.
Become an email marketing guru at the GURU conference. It’s two days full of all things email marketing. Learn more about newsletters, deliverability, design trends, AI, and what NOT to do with email.
What you can expect:
Keynote Speakers: Nicole Kidman, Amy Porterfield & more!
The latest digital trends in email marketing & how to increase performance.
Networking opportunities - each day!
Dj’s, dance contests (judged by Lance Bass, yes for real), breaking world records & MORE!
Spots are limited. It’s VIRTUAL. It’s FREE. It’s time to become an email marketing GURU. Join 25,000+ marketers on November 6th & 7th. Don’t miss out!
AI Tools of the Day
🎮 Bitmagic - Instantly turn your game ideas into playable 3D worlds with multiplayer support, all without writing a single line of code.
👥 Tavus - Scale personalized outreach by transforming a single recording into thousands of unique AI-generated videos with your own face and voice.
🕸️ ScrapeGraphAI - Use simple natural language instructions to scrape any website and convert messy web pages into clean, structured data for your AI agents.
⚙️ Toolhouse - Deploy production-ready AI agents and complex workflows in minutes using a universal SDK and pre-built cloud infrastructure.
🧠 VidText AI - Instantly transform lectures, meetings, or videos into accurate text transcripts, summaries, and structured mind maps for better understanding.
🎤 AI Clone Voice Free - Create a realistic digital replica of your own voice from just a few seconds of audio to generate expressive text-to-speech content.
📱 a0.dev - Build and launch fully functional native mobile apps simply by describing what you want your application to do.
🎨 GoEnhance AI - Animate characters, swap faces, and transform your ordinary videos into extraordinary moving art with powerful video-to-video AI models.
Explore the Best AI Tools Directory to find tools that will 10x your output 📈
China's Nvidia Lockout

The Recap: In a significant escalation of the US-China tech rivalry, Beijing has reportedly ordered its top tech firms to halt all purchases of Nvidia's AI chips. The directive affects specialized chips designed for the Chinese market and marks a major push toward domestic alternatives.
Unpacked:
The order from China's Cyberspace Administration targets the RTX Pro 6000D, a chip Nvidia designed explicitly to navigate U.S. export controls.
This directive aligns with Beijing's push for technological self-sufficiency, as domestic players like Alibaba secure major deals for their homegrown AI chips.
Nvidia’s stock fell nearly 3% following the news, with CEO Jensen Huang stating he was "disappointed" but patient amid the unfolding geopolitical tensions.
Bottom line: This move effectively walls off a massive market for Nvidia and accelerates the creation of a separate Chinese tech ecosystem powered by local hardware. For tech professionals, this signals a future with two distinct, and potentially incompatible, AI development environments.
AI Training
The Recap: In this video, I'm gonna break down a recent deal we just closed with a law firm that specializes in insurance mediation. We're going to show you the full sales process we followed to source and close this deal, along with a detailed technical breakdown of the AI automation workflow we created to source new leads for this law firm.
P.S. We also launched a free AI Automation Community for those looking to build and sell AI Automations — Come join us!
Groq's $750M War Chest

The Recap: AI chip startup Groq pulled in a massive $750 million from investors like BlackRock and Samsung, boosting its valuation to $6.9 billion to take on Nvidia in the AI inference race.
Unpacked:
The new capital will directly fund an expansion of data center capacity, with an Asia-Pacific facility planned for 2025 to meet customer demand that already outstrips supply.
While Nvidia dominates the AI training market, Groq is carving out a niche by focusing on inference—the process of running AI models in real-world applications.
Groq sells compute power bundled with its custom silicon and is already a key infrastructure partner for Saudi Arabia's Humain, signaling early traction in international markets.
Bottom line: This deal highlights the intense investor search for credible alternatives to the current AI chip monopoly. Increased competition in the inference space ultimately means more options and potentially lower costs for developers building AI-powered products.
Where AI Experts Share Their Best Work
Join our Free AI Automation Community
Join our FREE community AI Automation Mastery — where entrepreneurs, AI builders, and AI agency owners share templates, solve problems together, and learn from each other's wins (and mistakes).
What makes our community different:
Real peer support from people building actual AI businesses
Complete access to download our automation library of battle-tested n8n templates
Collaborate and problem-solve with AI experts when you get stuck
Dive into our course materials, collaborate with experienced builders, and turn automation challenges into shared wins. Join here (completely free).
An AI Health Forecast

The Recap: Researchers have developed a new generative AI model, Delphi-2M, that can forecast a person’s risk for over 1,000 diseases up to a decade in advance. The model, detailed in a new Nature paper, analyzes patterns in medical history to predict future health outcomes.
Unpacked:
The model works like a large language model for health, learning the "grammar" of disease by analyzing sequences of diagnoses, lifestyle factors, and other events from patient records.
It was trained on anonymized data from over 400,000 people in the UK Biobank research project and then successfully validated on 1.9 million patient records in Denmark.
Instead of giving exact predictions, the AI provides probabilistic forecasts, similar to a weather report estimating a 70% chance of rain for your future health.
Bottom line: This technology points toward a future where healthcare can shift from reactive treatment to proactive, personalized prevention based on individual risk profiles. Generative models like Delphi-2M could help clinicians identify high-risk patients earlier and allow health systems to better anticipate future needs.
The Shortlist
Anthropic reportedly clashed with the White House over its refusal to allow its AI models to be used for certain law enforcement surveillance purposes, deepening tensions with the Trump administration.
Amazon launched a new agentic AI tool in its Creative Studio that can generate ideas, storyboards, and complete video ads, aiming to reduce production time from weeks to hours.
SMIC is testing China's first domestically-built advanced DUV lithography machine, a major step in the country's push for semiconductor self-sufficiency for AI processors.
Reddit is reportedly discussing a new, more dynamic AI content deal with Google that would tie payments to how vital its data becomes for AI-generated answers.
Fiverr laid off approximately 30% of its staff, or 250 employees, as the freelance marketplace pivots to become an "AI-first company" with a flatter organizational structure.
What did you think of today's email?Before you go we’d love to know what you thought of today's newsletter. We read every single message to help improve The Recap experience. |
Signing off,
David, Lucas, Mitchell — The Recap editorial team