Tom Grubb’s Post

Huge thanks to @alexalbert for sharing a sneak peek of Claude 3! I'm particularly excited about Claude FM, which promises to: Reduce prompt engineering time – This is a game-changer for efficiency! Improve agentic workflow – Claude 3 seems to be taking big strides in automation. Extract features from images – Multimodal capabilities open exciting new avenues. All this while maintaining a high bar for responsible AI development. Additionally, the potential of using Haiku as a sub-agent (as explored in https://github.com/haiku) is intriguing. This is definitely a development to keep an eye on! #Claude3 #AI #MachineLearning #ResponsibleAI

"How do I easily reskin my game asset with AI? Whilst keeping the camera angle." Easy. Head to Scenario. 1. Select a platform model, in this case I picked "Classic 3D Realism". 2. Input a Reference Image, in this case an image of the exact treasure chest I want. 3. Use ControlNet Depth from the drop down. 4. Prompt the type of render or theme you'd like your treasure chest to have. Here is the loom I recorded for the full process: https://lnkd.in/dspeJQgm The whole process took me 6 mins. Happy creating. Suggest Workflows below👇 #gamedev #gameart

As Ed rightly points out, this is a great summary of how someone is using different AIs and how in their own work. Worth reading through for a practical example of how AI can be useful, and how some of the idiosyncrasies of AI can be beneficial if used or regarded in the right way. But throughout the presentation I'm very conscious of how much work Zoe Scaman is putting into using a tool - I wonder if LLMs aren't inherently that much better than traditional methods, but the interface is just a lot more interesting and interactive? And that just particularly suits the area Zoe is in? ( and as a side note, it was "interesting" to see, in such a thoughtful and advanced field, that using AI as a Devil's Advocate was seen as odd, and not standard practice )

We are in the age of Multimodal RAG and Knowledge Graphs...if you said "What's that"...you are already lagging behind. Check these videos https://lnkd.in/esbbVxWi https://lnkd.in/e3AXQ9kn

For all my Strategy friends out there- I found this presentation by Zoe Scaman to be the most helpful and inspiring guide yet on how we can use AI to make our work and ourselves smarter, prettier, and more impactful. Do not miss it- and please let me know if there are any tools you're using that I might not know about yet, or that she doesn't reference herein. https://lnkd.in/gchk3XGn

🚀 Excited to share that I have completed the "Knowledge Graphs for RAG" course organized by Neo4j and DeepLearning.AI! 🌐📊 This knowledge opens up new possibilities in the integration of knowledge graphs with Retrieval-Augmented Generation (RAG), enhancing the way we process and generate information in AI and ML. 💡 A big thanks to the organizers for an incredible learning experience!

A quite comprehensive & pragmatic guide for strategists exploring efficiency & pertinence through the AI hype. Such an insightful read by Zoe Scaman. Thanks Edward Cotton for sharing. #strategy #AI

Florence-2, an innovative vision-language model by Microsoft. This cutting-edge model showcases robust zero-shot and fine-tuning capabilities across a variety of tasks, including captioning, object detection, grounding, and segmentation. Florence-2 Paper explanation: https://lnkd.in/gQ75tVJn Try out the Florence-2 model here: https://lnkd.in/g9Cudu4k Paper: https://lnkd.in/gpSbjafa #computervision #largelanguagemodels #languagemodels #microsoft #ai #artificialintelligence

It's about that time for another Agent700 tidbit. In this session, we'll show you how to use a transcript to evaluate emotional intelligence. https://lnkd.in/gPH6eQGs

Imagine if you could hold a 𝐭𝐡𝐨𝐮𝐠𝐡𝐭𝐟𝐮𝐥 𝐜𝐨𝐧𝐯𝐞𝐫𝐬𝐚𝐭𝐢𝐨𝐧 with 250 people across your company in real-time, brainstorming and deliberating and prioritizing as a unified Collective Intelligence that finds optimal solutions... Or, imagine holding a POLICY BRAINSTORM with 400 members of the public in real-time, enabling the online group discuss and debate regulatory ideas so you can quickly surface solutions that satisfy diverse perspective... A new technology called Conversational Swarm Intelligence (CSI) now makes these type of large-scale conversations possible. It combines the power of biological swarms with the flexibility of Generative AI. And, it works... See a recording of my KEYNOTE from Collective Intelligence 2024 below for a review of the science, the technology, and a new tool called Thinkscape that makes this possible for large groups around the world. For more info on Thinkscape: https://www.thinkscape.ai/ To watch the Keynote: https://lnkd.in/grJkphcv David Baltaxe Joshua Sitzer Albert Lim Gregg Willcox Nichol Bradford Ganesh Mani Future of Marketing Institute Lifeboat Foundation AI for Good GatherVerse XRSI Unanimous AI #collectiveintelligence #civicengagement #deliberativedemocracy #swarmintelligence #swarmAI #thinkscape #collectivesuperintelligence