Sochin Research Institute’s Post

"...the computer scientists come in and say, look, we're going to make intelligence just with next token prediction and gradient descent and transformers. And then they say, but you know, this intelligence would be much better if we just had some more data from people that we could add to it. And then all the villagers go out and add all of the data of everything that they've uploaded to the internet. And then the computer scientists say, no, this is doing a good job at being intelligent. But it would be even better if we could have reinforcement learning from human feedback and get all you humans to tell it what you think is intelligent or not. And all the humans say, OK, we'll do that. And then and then it would say, you know, this is really good. We've got a lot of intelligence here. But it would be even better if the humans could do prompt engineering to decide exactly how they were going to ask the questions so that the systems could do intelligent answers. And then at the end of that, the computer scientists would say, see, we got intelligence just with our algorithms. We didn't have to depend on anything else. I think that's a pretty good metaphor for what's happened in AI recently."

Tunadorable is a YT personality who does AI-related deep-dives. Here's an interesting one: https://lnkd.in/emh3A37P The short answer is: no, we're not even close to understanding LLMs. But there are a few annoying oversimplifications about new AI tech we can get rid of now! Here are two cheap shots I've heard a lot of lately: "LLMs are just stochastic parrots" or "LLMs are just glorified next-token predictors." Well, technically, yes. Just like "a brand new functional iPhone and this pile of unassembled phone parts are just the same set of parts, so are functionally equivalent." That is completely missing the point. Learn the bitter lesson: when you do something boring, but do it bigger, the emergent properties of the system are surprising and novel. Those of us who've been building massive clusters over the past 30 years know this intuitively. Examples? You can MODIFY LLMs to be more (or less!) honest, and they'll do it. You can give them code with an error and make them unable to accept code can have errors, and they will go so far as to rewrite the code error-free in order to resolve the conflict. These behaviours are novel and interesting and can teach us a few things about ourselves.

Attended this interesting session on AI and human intelligence.

God is already dead, now truth too.... By the latter half of the 20th century, some argued that we were entering a post-truth era, where Truth—understood as a universal, objective reality—was replaced by truths, more contextual and transient. This shift was driven by the subaltern deconstruction of dominant narratives, which questioned the legitimacy of mainstream hegemonic perspectives. The advent of the World Wide Web marked a true information revolution, giving unprecedented access to scientific reports, news, legal documents, and more. While it opened space for alternative viewpoints, it also expanded access to what was considered Truth. However, today’s AI is blurring the line between socially sanctioned Truths—such as academic research, court rulings, and constitutions—and truths that gain prominence based on entertainment value and social validation. What happens when we begin to substitute Truth with Likeability?

🎉 The 8 Types of Intelligence: Understanding the Full Spectrum of Human Potential 🎉  ***************************************** We’re still buzzing with excitement from our latest BrAIN event, where some of the brightest minds in AI came together to share insights, discuss advancements, and inspire innovation. For those who couldn’t attend—or if you want to revisit the incredible discussions—we’ve created a special video-podcast featuring the event’s most important highlights! Why a video-podcast? We wanted to make this as engaging as possible by integrating AI-generated visuals into the discussion. These dynamic visuals were inspired by the event’s themes and showcase the incredible diversity of thought and innovation within our community. In this video-podcast, you’ll find discussions related to Howard Gardner’s Theory of Multiple Intelligences and how AI intersects with various forms of intelligence: 1. Spatial Intelligence – Enhancing AI’s ability to process spatial and visual data, with applications in design and autonomous systems. 2. Kinesthetic Intelligence – Exploring AI in robotics and its role in precise motor control and human-robot interactions. 3. Musical Intelligence – AI-generated music and how AI tools like deep learning models are revolutionizing composition and sound analysis. 4. Linguistic Intelligence – The latest advancements in large language models (LLMs) and their transformative applications. 5. Logical-Mathematical Intelligence – Discussions on AI’s problem-solving capabilities and computational power in fields like finance and engineering. 6. Interpersonal Intelligence – Using AI to enhance communication, from customer service bots to collaboration tools. 7. Intrapersonal Intelligence – Insights into how AI helps individuals understand and optimize their emotions and mental well-being. 8. Naturalistic Intelligence – Exploring how AI analyzes and predicts patterns in nature, contributing to sustainability and conservation efforts. Whether you’re a seasoned AI expert or just beginning your journey, this video-podcast has something for everyone. The conversations were rich, thought-provoking, and a true testament to the power of collaboration within our community. 🎥 Watch the podcast here: https://lnkd.in/giu89s5C We’d love to hear your thoughts! Which parts stood out to you? Let’s keep the conversation going—your feedback and insights are invaluable to shaping our future events. Thank you to everyone who attended, contributed, and made this event a success. Stay tuned for updates on our next gathering. Until then, enjoy the podcast! Sivant M, Richali Shah, Sima Yazdani, PhD, bereket issack, Atsushi Hisaka, Chris Harget, DB Biswas, Sagar Phadkule, Lonnie Chrisman, Sam Mirazi, Łukasz Milewski, Esfandiar Bandari, Arka Bagchi, Karandeep Singh #BrAIN🧠 #AI #ArtificialIntelligence #LLM #AIPodcast #MachineLearning #AIEvent #BayAreaTech

I really like this article - worth a read. Covers different angles of #artificialintelligence

Or what we call in modern AI - a hallucination.

It is Peer Review Week. As our industry across the globe focus on the impact of innovation and technology (AI) on the peer-review process, we consider the impact of bad manners

With the consistant #ethical concerns around #ai, thay occur on a daily basis. And with AI development embracing #humanemotions (#emotionAI) e.g. the company Hume AI, the industry and VCs are throwing caution to the wind about the implications of this. This only hightens concerns for the potential impact on society, especialy when #AI already does not “design for multi-cultural, intersectional possibilities”. For more about this topic, and beyond, the incredible Suhair Khan explores this is the latest ‘the futures of intelligence’ post, below. #ai #emotionai #inclusiveai #accountablity #responsibleai #hcai #context Alexandra Deschamps-Sonsino Melanie Smith Stephanie Hare Alex Ash Kyle Soo Delfina Fantini van Ditmar Seán Boyle Myles Igwebuike https://lnkd.in/eRjW_kRb

Evolutionary Theory vs. Intelligent Design and Syntropy chatting with claude3 Opus - see the conversation at aiyadda.com