| |
Good morning, {{ first_name | AI enthusiasts }}.The latest AI breakthrough isn’t coming from a tech giant — it’s emerging from a university lab with a shoestring budget.
UC Berkeley just matched o1-preview’s capabilities with a $450 training budget in 19 hours, and an open-source reasoning revolution may be right around the corner.
In today’s AI rundown:
-
$450 open-source reasoning model matches o1
-
OpenAI building out new robotics division
-
Design your personal habit transformation roadmap
-
World Economic Forum forecasts AI workplace surge
-
4 new AI tools & 4 job opportunities
LATEST DEVELOPMENTS
UC BERKELEY
🧠
Image source: UC Berkeley
The Rundown: UC Berkeley’s NovaSky team just Sky-T1-32B-Preview, a fully open-source reasoning model that competes with earlier versions of OpenAI’s o1 despite minuscule training time and costs.
The details:
-
Sky-T1 is a fine-tuned version of Alibaba’s Qwen2.5-32-Instruct, with training data generated using the open-source reasoning model QwQ-32B-Preview.
-
Training took just 19 hours on 8 H100 GPUs, and the total cost was around $450 — a fraction of typical AI training budgets.
-
The model matches or exceeds an earlier version of OpenAI’s o1 on several benchmarks, particularly excelling in mathematics and coding challenges.
-
Unlike other reasoning models, Sky-T1’s entire pipeline, including training data, code, and model weights, is completely open source.
Why it matters: Open-source AI has hit yet another milestone — with UC Berkeley showing that high-level reasoning can be replicated at a fraction of the cost and training time of the massive AI giants. A new wave of innovation could come from previously priced-out labs that can now train and develop reasoning models.
TOGETHER WITH PROJECT IDX
☁️
The Rundown Project IDX gives you a complete cloud development environment with Google’s Gemini AI, right in your browser.
With Project IDX, you get:
-
Your laptop in the cloud: A Linux-based virtual machine
-
Native apps in browser: Build Flutter/React Native apps with a built-in Android Emulator
-
Pre-loaded environments: One click to start with popular frameworks
.
OPENAI
🤖
Image source: Midjourney / The Rundown
The Rundown: OpenAI just the company’s first robotics hardware job listings, revealing plans to develop its own custom robots with advanced AI capabilities.
The details:
-
Former Meta AR glasses lead Caitlin Kalinowski is spearheading the effort, joining as OpenAI’s hardware director in November.
-
The company is hiring for technical roles, including sensor suite development and mechanical design, and a lab operations manager to oversee prototype testing.
-
Job listings hint at for ‘general-purpose robots that operate in dynamic real-world settings,’ with plans for a ‘wide variety of robotic form factors.’
-
OpenAI shuttered the robotics team in 2020, with research including a robotic hand to solve a Rubik’s Cube and other challenges.
-
OpenAI has also with Figure in the past year, integrating its models into the robotics firm’s humanoid robots.
Why it matters: While OpenAI is no stranger to robotics hardware with its partnerships and reported consumer device efforts with Jony Ive, rebuilding an in-house robotics division may signal a belief that achieving its AGI goal may require control of both the physical and digital aspects of AI systems.
AI TRAINING
🎯
The Rundown: Transform your habits by creating a personalized development plan that uses AI to combine expert knowledge with your specific situation.
Step-by-step:
-
Access any conversational AI assistant (e.g., ChatGPT, Claude, or Gemini).
-
Share your context using the profile prompt to get personalized questions about your goals and obstacles.
-
Request expert insights with the knowledge base prompt to get evidence-based strategies.
-
Generate your custom action plan with the final prompt that combines your situation with expert guidance.
Pro tip: Keep your conversation saved as a template for developing future habits! We have also written detailed prompts that you can copy and paste .
PRESENTED BY AMARO
💡
The Rundown: Amaro is your all-in-one platform for AI content creation, empowering enterprises and creatives to leverage the latest AI models across image, video, audio, and text.
With Amaro, you can:
-
Create brand-consistent content using custom AI models trained on your assets
-
Switch between leading AI tools like Runway, FLUX, and ElevenLabs in one intuitive workspace
-
Generate production-level product photos with customizable settings and backgrounds
to see how Amaro makes AI work for you.
AI & THE WORKFORCE
📈
Image source: World Economic Forum
The Rundown: The World Economic Forum just its 2025 Future of Jobs Report, revealing AI’s dramatic impact on the workforce — with an overwhelming majority of employers expecting significant transformations from the tech by 2030.
The details:
-
Technology adoption is surging, with 86% of companies expecting AI to transform their operations by 2030.
-
AI is predicted to create 11M jobs while displacing 9M others, with big data specialists and AI/ML experts topping the list of fastest-growing roles globally.
-
Three-quarters of organizations plan to upskill existing employees for AI collaboration, while 70% aim to hire new staff with AI experience.
-
Half of companies expect to reorient their business around AI opportunities, while 40% anticipate reducing workforce size as AI capabilities grow.
Why it matters: AI’s disruption to the workforce is coming fast, and every industry should be planning its talent and tech strategies to prepare for the massive changes ahead. Early adopters who successfully navigate the AI boom will see major competitive advantages during modern history’s biggest reshaping of work.
QUICK HITS
🛠️
-
🤖 – Create powerful AI agents with minimal effort
-
📓 – Reduce anxiety with AI-powered guided journaling
-
👨💻 – Github + Pinterest to make your AI websites look beautiful
-
🪞 – Showcase products with AI avatars
💼
-
📦 – General Application, Data Operations Engineer
-
💡 – Patent Agent
-
📈 – Business Operations Analyst
-
🗓️ – Executive Assistant
📰
Elon Musk during an X interview at CES 2025 that AI has exhausted the sum of human training data, and companies are turning to AI-generated synthetic data to train future models despite the drawbacks.
Nvidia an AI Blueprint for retail shopping assistants, enabling AI-powered digital retail agents to process text and image queries while visualizing products in virtual spaces.
AMD researchers Agent Laboratory, a framework for using LLM agents as research assistants that can handle literature reviews, experiments, and report writing — achieving an 84% cost reduction compared to previous methods.
Mark Zuckerberg in an interview with Joe Rogan that Meta and other companies will use AI to automate the roles of midlevel engineers this year, with plans to offload all coding to artificial intelligence eventually.
Savannah Feder Astral, a viral automated AI marketing platform that can handle social media tasks like commenting and content creation on Reddit.
A new report from Bloomberg Intelligence that over 200,000 Wall Street jobs may be slashed in the next 3-5 years due to AI, with automation and productivity increases leading to up to 17% higher banking profits.
COMMUNITY
🎥
Join our next workshop on Wednesday at 2 PM EST and learn how to create custom styles for brand consistency via Freepik with Sofía López Martín.
RSVP . Not a member yet? Join on a 14-day free trial.
🤝
We’ll always keep this newsletter 100% free. To support our work, consider with your friends, and we’ll send you more free goodies.
Your current referral count: {{rp_num_referrals}}
What you need: {{rp_num_referrals_until_next_milestone}} more referrals until receiving {{rp_next_milestone_name}}
See you soon,
Rowan, Joey, Zach, Alvaro, and Jason—The Rundown’s editorial team