Lukas Mosser’s Post

I look forward to taking o1 for a spin via API and ChatGPT Plus. Models being better at reasoning is one of the barriers to break and the next step change in utility and value creation 👀 “This new series of AI models can reason through complex tasks and solve harder problems than previous models in science, coding, and math. “

🚨 Exciting News from OpenAI! 🚨 OpenAI has unveiled a new model, previously codenamed "Strawberry," now officially called OpenAI o1 ! 🍓 💡 Why it matters: This model evaluates its steps before proceeding, making it better at handling complex tasks like math, science, and coding. Plus, it’s more explainable and safer! ✅ 📢 Key Highlights:   - A lightweight version, o1-mini, is designed for code generation.   - Gradual rollout: ChatGPT Plus & Team users get early access today, with more coming soon.   - Weekly message limits apply (30 for o1-preview, 50 for o1-mini).  ⚠️ Limitations: It’s text-only, slower, and can't reason with documents or fetch real-time info (yet). ⏳ #AI #OpenAI #artificialintelligence

And it just keeps getting better!! #OpenAI Strawberry (this is the version name of the new OpenAI) 🍓🍓🍓🍓🍓🍓🍓🍓🍓🍓🍓🍓 OpenAI, #ChatGPT, and #Copilot bring me back to my live stream days. Live stream empowered me to build a brand and a company that changed my life. It gave me the edge I needed to break into a "new thing" quickly. Now Generative AI is doing the same for so many people. #GenAI is leveling the playing field for creative, technology and business professionals. - Where you went to school? Doesn't matter. - Who you know? Doesn't matter. - Where you grew up? Doesn't matter. Learn #AI #prompting (how to chat with AI) now and incorporate it into your day-to-day workflows.... I promise it's worth it. How do you use AI? #productivity #AIatWork

Wow. It’s getting deeper and deeper. We saw that in. 1980 as the computer power increases. Now it happens again. But more in the direction of AI. The next generation of AI models are designed to [think] more before they answer.

📢 Breaking News: OpenAI Unveils o1 Preview! 🧠💡 OpenAI announced the release of their groundbreaking o1 series - AI models designed to think longer and deeper before responding. Key highlights: 🚀 Available now for ChatGPT Plus and Team users, and API developers (tier 5) 🧪 Excels in complex reasoning tasks for science, coding, and math 🏆 Outperforms PhD students in physics, chemistry, and biology benchmarks 💻 Reaches 89th percentile in Codeforces coding competitions 🥇 Scores 83% on International Mathematics Olympiad qualifying exam (vs. 13% for GPT-4) 🛡️ Enhanced safety features: New training approach leveraging reasoning capabilities Improved resistance to "jailbreaking" attempts (o1-preview scored 84/100 vs. GPT-4's 22/100) Rigorous testing using Preparedness Framework Best-in-class red teaming and board-level review processes 🌐 Government collaboration: Formalized agreements with U.S. and U.K. AI Safety Institutes Early access granted to research version for evaluation and testing 👥 Potential applications: Healthcare research (e.g., annotating cell sequencing data) Quantum physics (generating complex mathematical formulas) Advanced software development (multi-step workflows) ⚙️ Technical insights: Uses large-scale reinforcement learning to enhance thinking processes Performance improves with more training compute and thinking time Exploring new scaling constraints different from LLM pretraining 🚧 Current limitations: Lacks some features of GPT-4 (web browsing, file/image uploads) Still in preview, with regular updates expected 💰 Pricing: $3.00 per 1M input tokens This release marks a significant leap in AI capabilities, potentially revolutionizing complex problem-solving across various fields. OpenAI is resetting the counter to 1 with this new series, signifying a new era in AI development. What potential impacts do you foresee from this advancement in AI reasoning capabilities? How might it affect your industry? Share your thoughts below! 👇 #OpenAI #AI

OpenAI just announced its new AI model series: o1. Unlike its debut of GPT-4o in May, which boasted its ability to reason across text, vision, and audio inputs quickly, OpenAI is optimizing o1 to spend more time thinking and reasoning to solve more complex problems. If you're looking for a quick sense of how this model series is improving on GPT-4o, OpenAI shared their benchmarking tests here: https://lnkd.in/e3eQvViM There were improvements in its performance across several fields (including #law, yay), but the most significant improvements were in STEM/CS categories. I'll need to do more tinkering with o1 before I can say much more, but I've done a few quick tests with old prompts. The sample responses were impressive improvements over GPT-4o. I'm excited about what we can build with these new models -- what are your first impressions of o1? #AI #artificialintelligence #o1 #emergingtech #legaltech

When I receive something like this, I'm immediately thinking about the awesome implications it can have on multiplying our intelectual capacity For example, the case study below. But coming from the humanities and having a lot of time invested in social-cultural issues and politics, there's a natural arising of an analysis of how different groups will view these developments. I share some huge concerns surrounding AI—ethical considerations, potential job disruptions, and the fear of technology diminishing our capabilities. This was strengthened by recent conversations with various friends from different cultural and political backgrounds, expressing fears about declines in creativity, AI warfare, and job destruction. What then? I believe though, it's equally important to highlight how AI can amplify human potential. By expanding our capacity to think, work, and connect, we can use these tools to foster growth rather than hinder it. I use AI to find out if the bugs in my room are trouble (I'm decently inside the forest) and playing with kids, I' ve also been having increasingly more impulse of writing manually in books, composing music, and working on paintings. Then while I do some trekking through the week I use GPT to transcribe my thoughts and help start some important messages or posts. These practices liberate time for me to engage in genuinely creative work. Ultimately, AI should progressively remove the non-creative aspects of work, allowing us to concentrate on what truly matters. I believe that as humans, we will select and preserve the creative elements. I do not think that AI-generated art and the creative process of AI will thrive. While AI-created art might exist in a limited capacity, the art that rises will be truly creative and authentically human. Even if AI mediates some aspects, without human creativity at its core, it won’t sustain or flourish. Next month it will be an honor to be on a panel on AI & Ethics at a global conference, where I hope to tackle some of these topics and explore other aspects of ethics around AI use, including how we discuss AI, what we focus on and the broader issues surrounding these topics. Our growing capabilities come with increased responsibilities—not only for ourselves but also for those around us and the communities we cherish. It's essential to be there for each other, for our families, and for those in need, ensuring that we navigate a path that benefits everyone. Relationships, whether personal or professional, will play a pivotal role in directing the future. Relying solely on government regulations might be insufficient; instead, fostering profound discussions rooted in shared values is crucial. A lot of the joy or pain we'll go through in the next few years will be heavily influenced by how we and those in power act. I hope we go on the path of human empowerment through AI, either through leadership decision or consumer choice. #openai #gpto1 #ethics #mindsets

🚀 Exciting Times for AI-Driven Branding! With the new ChatGPT preview, we’re entering a groundbreaking era for brands. Here’s what we’re most excited about at BrandLab AI: 1. Hyper-Personalization: AI-powered, tailored customer experiences at scale—revolutionizing how brands connect. 2. AI-Enhanced Creativity: From content generation to campaign ideas, ChatGPT empowers creative processes like never before. 3. Deeper Customer Engagement: Smarter, real-time conversational interactions to boost brand responsiveness. 4. Data-Driven Insights: Powerful analytics that drive informed, agile brand strategies. At BrandLab AI, we’re ready to help brands harness these tools to create innovative, sustainable, and ethical branding solutions. #AI #branding #innovation #OpenAI #ChatGPT #personalization #sustainability

Impressed with how OpenAI managed to use reinforcement learning for chain of though in the new o1 reasoning model. Here is an experiment I run, where o1 created a whole machine learning model in 1 shot, Im impressed with the results. https://lnkd.in/eCxhAFUi