Suno really is incredible. Made a full album in an evening with music I actually enjoy listening to while working. On repeat🤯 You can check out the album here: suno.com/playlist/f049d…

Mark the day, as this Thursday we will share some more info about AI People! 🥳 📻Stay tuned and join the AI People revolution now! aipeoplegame.com #AIPeople #GoodAI

AI People Alpha Launch: AI NPCs Beg Players Not to Turn Off the Game! Check out our Alpha Release Trailer! 👀 youtu.be/3XtqQBt-W0Y Join the revolution and Waitlist AI People now! 💡 waitinglist.aipeoplegame.com #AIPeople #AlphaTrailer #AIRevolution

🎉The wait is over! AI People has finally reached the public Alpha stage! 🎮✨ Step into a world where AI NPCs think, feel, and are alive! Be part of the revolution and experience the future of gaming! 🚀 ➡️ goodai.com/ai-people-alph… ❗ Important ❗ The access will be granted in

AI People is finally available in public alpha! 👍😀 Access will be rolled out in phases through a waiting list system.

Our paper 'Beyond Prompts: Dynamic Conversational Benchmarking of Large Language Models' has been accepted to 2024 Conference on Neural Information Processing Systems! arxiv.org/abs/2409.20222 Join us at our poster session to chat and explore our work on Long-Term Memory and

📣 Update 0.1.11a is live! 📣 We are excited to announce that today we have released our first content addition update! Now you can talk to our agents with your own voice through a microphone, eliminating the need to type everything! Just press and hold the left mouse button

Check the 🔄Word Swap Challenge🔄: for just a bunch of cents, you can test an LLM and predict how good it will be at understanding your prompts. Apart from the great results, I really enjoyed collaborating with Jan Feyereisl on this work.

Read more about the Word Swap Challenge in the GoodAI blog goodai.com/breaking-the-c…

Fascinating new LLM benchmark alert! 🚨 My colleagues David Castillo and Jan Feyereisl have developed a simple yet powerful test exposing LLMs' struggles with sequential reasoning. Key findings: • Most LLMs start erring after just 2 operations 😮 • OpenAI o1-mini unsurprisingly

This sounds very much related to the Word Swap Challenge. Claude was quite good at it. Looking forward to test the new version. github.com/GoodAI/llm-cha…

🔔 Update 0.2.0a coming tomorrow! 🔔 🪑 The Sitting Pretty Update brings plenty of new content to AI People! You can look forward for Update stream: 🕖 When: Thursday 24/10/2024 at 19:00. 📌 Where: Twitch twitch.tv/aipeoplegame/ and YouTube youtube.com/@GoodAI/ Check our

🚨 The Sitting Pretty Update of AI People is live 🚨 Brand new AI People update is live and running! Read more info on our website: goodai.com/ai-people-the-… 👀 We would like to invite you to watch our AI People Update stream today from 19:00 CEST (UTC+2) on: 🟣 Twitch:

Next week, we’re releasing a major update for AI People: Local LLM. Our AI NPCs will be able to run entirely on your PC, eliminating the need for cloud-based LLMs. To my knowledge, this makes us the first game in the history of the universe to achieve this! 💪

This Thursday at 19:00 CET - join us for the biggest Space Engineers 2 reveal stream in a decade! Get ready for new SE2 trailer, roadmap, live gameplay, and more. I am grateful to our incredible team - working meticulously, pushing the quality, and showing true dedication. Best

8. AI People: NPCs interact dynamically and evolve stories.

🎮 GoodAI is revolutionizing NPC interaction in their new game, AI People, where you can experience AI-driven characters that learn, feel, and create dynamic narratives. #CES2025

After some thought, I can elaborate why I'm mind-blown: It's now provably possible to train a model to execute a computation (Bellman-Ford in this case) and actually get *guarantees* that the model learnt that computation (up to training error) and not a bag of weak heuristics.