OpenAI announces new o3 models

Start learning AI in 2025

Everyone talks about AI, but no one has the time to learn it. So, we found the easiest way to learn AI in as little time as possible: The Rundown AI.

It's a free AI newsletter that keeps you up-to-date on the latest AI news, and teaches you how to apply it in just 5 minutes a day.

Plus, complete the quiz after signing up and they’ll recommend the best AI tools, guides, and courses – tailored to your needs.

Sign up to start learning.

🔥 OpenAI announces new o3 models - LINK

• OpenAI has introduced the o3 model, skipping the o2 version due to copyright issues, and it showcases significant advancements in reasoning capabilities, outperforming its predecessor, o1, by a substantial margin in various benchmarks.

• The o3 model achieved a remarkable 25.2% success rate on complex math and reasoning problems, where previous models struggled to exceed 2%, and it scored nearly perfectly on the AIME 2024 math competition.

• OpenAI plans to launch o3-mini by the end of January 2024, which will offer different reasoning options and is part of a broader strategy to allow external safety assessments of its models.

☎️ OpenAI brings ChatGPT to your landline - LINK

• OpenAI has launched ChatGPT for landlines, allowing users to call 1-800-CHATGPT (1-800-242-8478) and interact with the AI-powered assistant, expanding accessibility beyond smartphones.

• The service mirrors the Advanced Voice Mode of ChatGPT, enabling real-time conversations where users can ask questions and receive responses, including tasks like translation.

• U.S. users will receive 15 minutes of free calling to try out the service, and ChatGPT is also accessible via WhatsApp for those who prefer text-based interactions.

🧠 Google releases its own ‘reasoning’ AI model - LINK

• Google has introduced the Gemini 2.0 Flash Thinking Experimental model, an experimental reasoning AI designed for multimodal understanding, reasoning, and coding, available through its AI prototyping platform, AI Studio.

• The model aims to enhance reasoning capabilities by effectively fact-checking itself, but it may take longer to provide answers, often pausing to consider related prompts before summarizing the best response.

• Despite promising results in initial testing, there are concerns regarding the model's performance consistency and the high computational costs associated with reasoning models, as they require significant resources to operate effectively.

✨ Instagram teases AI tools for editing appearances - LINK

• Instagram is set to introduce generative AI tools that will enable creators to modify various aspects of their videos using text prompts, allowing for changes in appearance, outfits, and backgrounds, as teased by head Adam Mosseri.

• The editing features will utilize Meta’s Movie Gen AI model, which allows users to create and edit videos through simple text inputs, although it remains uncertain if the final user-facing tools will achieve the same seamless results shown in previews.

• This announcement follows the trend of similar AI models launched by competitors, including OpenAI's Sora and Adobe's Firefly video generator, indicating a growing focus on AI-driven content creation tools across social media platforms.

💰 Perplexity Bags $500M Funding - LINK

• Perplexity has successfully closed a $500 million funding round, bringing its valuation to $9 billion, with the investment led by Institutional Venture Partners and finalized earlier in December.

• The funding comes amid increasing competition in the AI-powered search sector, particularly with OpenAI's recent launch of ChatGPT Search and Google's plans to enhance its search capabilities with AI-generated content and a potential chatbot feature.

• In addition to securing funding, Perplexity has made its first acquisition by purchasing Carbon, a startup focused on integrating AI systems with external data sources, despite facing a class action lawsuit over alleged copyright infringement.

• 🚨 Tesla’s Big Recall: 700K Cars Over Tire Alerts Fail- LINK

• 👮‍♂️ Cops Crack Down as Amazon Warehouse Strikes Heat Up - LINK

• 🔓 Massive Data Leak Hits 240K Credit Union Members - LINK

• 🔑 Microsoft Wants You to Say Bye to Passwords - LINK

• 🚗 California Set to Ban New Gas Cars by 2035 - LINK

📖 How to Speak Machine

How to Speak Machine by John Maeda is a must-read for anyone curious about how technology and design work together. Maeda breaks down complex ideas into simple, relatable language, making it easy to follow even if you’re not a tech expert. He explains how machines “think” and why understanding their rules can help us design better, smarter products. The book is short, engaging, and packed with real-world examples. Perfect for creatives, designers, or anyone living in our digital world. It’s like a crash course in speaking the language of machines—fun, insightful, and surprisingly human