Weekly Link Roundup #50
Well those first 50 kind of flew by...
This marks the 50th Weekly Link Roundup from Relentlessly Curious and Deeply Engaged. It's a big one. I owe a huge thank you to anyone reading this or subscribing or sharing. I've enjoyed writing longer form items again, would probably do it if no one read it but I'd be lying if I said knowing that someone is reading it doesn't make it a little sweeter - so thank you all and here's to the next 50.
This falls under the heading of Breaking News as far as this newsletter goes - if you pay attention to the AI space, then you’ve seen the news about DeepSeek, the new model from China. It has literally scared the pants off lots of AI players and the stock market. There will be more to come in the minutes, hours, and days ahead but my thinking is this - DeepSeek has demonstrated a new model at orders of magnitude lower cost and whether or not its them or someone else, they’ve shown that the large incumbents are vulnerable. Also, I probably would not use them right now and certainly not for any kind of enterprise use. Read on.
Stunning breakthroughs from China's DeepSeek AI alarm U.S. rivals
DeepSeek R1’s bold bet on reinforcement learning: How it outpaced OpenAI at 3% of the cost
The fall of click-next e-learning: What Operator means for training: Ben Betts wrote a great piece about an experiment he did with OpenAI’s new mode Operator. His two most important lines? “I spent my weekend exploring the results, and honestly, it’s both brilliant and terrifying.” and “At the time, I thought, when they get this right, it could spell the end of e-learning as we know it.” If that coupled with the headline doesn’t get you to read the article, I don’t know what will.
See also: OpenAI’s new Operator AI agent can do things on the web for you
Tasks Versus Skills Part 1: Squaring the Circle of Work with AI: An important read from Marc Steven Ramos - “A 6-part position paper and eBook on the state of tasks versus skills, a playbook, and some provocative what-if concepts for task-enabled Learning and Development, and Talent teams.” See also: Reskilling in the Age of AI (HBR)
Love these fully realized scenarios. Even if you don’t agree with them, the envisioning is powerful and compelling. This is one possible agentic world.
Coevolution of AI and society: Study explores opportunities and risks: This is exactly the kind of study we need if we are going to better understand how our new tools will shape us. “The study, published by an international team of AI and complexity scientists in the journal Artificial Intelligence, titled "Human-AI Coevolution," explores how processes in a world where algorithms and recommendation systems increasingly guide daily decisions influence human behavior.”
Timex is making a wearable with a sensor to track brains, not hearts: This is one are, the data being produced by wearables, where I think there is room for good and ill. I think the thing that will make the difference is our understanding of how the data is collected and translated and where its strengths and weaknesses are. I think the secondary issue will be providing people with healthy strategies for making use of the data. So not just a sleep score but how to improve your sleep score. “Pison’s electroneurography (ENG) platform measures physiological electricity originating from your brain using a skin biosensor, which when combined with AI-powered software algorithms, can provide insights into mental health, sleep, sports performance, and brain health.”
I originally posted this on LinkedIn and it got some attention so I thought I’d include it here too with an additional link to a great HBR article on getting to a change mindset: OK...really noticing something...seeing a lot of articles on the "TRANSFORMATION" of work by integrating AI. That's fine and all but here's the bit I'm noticing...almost all the talk I've seen (and there I'm being kind, it's all the talk) is about how jobs will change, and how work will change, and what tasks will be automated, and what efficiencies will be gained. The part that's missing is all the rest...like how will we look at paying people differently? How will we change how we rate and assess performance? I mean if this is truly a transformation, how transformative can it be if we don't get to the whole system? Look, this examination of jobs and tasks is fine but we're going to do just that and then we'll wonder why people resist the change and we'll drag out some tired old adage like "people just resist change" - when we always mean that people resist a change that hasn't been explained or that they don't see the benefit in or worse, they see danger in. So can we maybe pause, take a step back, and look at how we can make foundational changes so that we create an atmosphere where, as April Rinne says, we move from change management to a change mindset? Thanks.
'Humanity's Last Exam' benchmark is stumping top AI models - can you do any better?: Sounds ominous doesn’t it? I’ll let you know how I do > > “On Thursday, Scale AI and the Center for AI Safety (CAIS) released Humanity's Last Exam (HLE), a new academic benchmark aiming to "test the limits of AI knowledge at the frontiers of human expertise," Scale AI said in a release. The test consists of 3,000 text and multi-modal questions on more than 100 subjects like math, science, and humanities, submitted by experts in a variety of fields.”
Why NotebookLM Matters: I agree with Om wholeheartedly on this > > “It’s a practical AI application, focusing on personal knowledge management rather than general-purpose chat. The system’s ability to process up to 750,000 words, expanding from 24,000, marks a significant leap in contextual understanding. This advancement allows users to derive insights from vast personal archives. NotebookLM’s innovation isn’t technical; it’s conceptual. The company is positioning AI as a tool for understanding rather than generating content. By doing so, Google can sidestep issues plaguing general-purpose AI while delivering genuine utility. Google should consider building more practical AI applications.”
Change Management Requires a Change Mindset by April Rinne: Just going to keep sharing this great article because I think it’s never been more important to stop looking at changes as something to be managed and look at it as a mindset to be adopted.
The Value of Knowledge Work in the Age of AI: In the spirit of sharing articles that I may have shared before but that I’ll share again because they’re important here we are again. I love this line “Whether you look at a corporate balance sheet or a homeowner’s insurance policy, there is no asset line item for people. Employees are only liabilities for organizations – their compensation and overhead (a.k.a fully burdened headcount).” Now think about what that means for L&D departments.
Pocket Worlds acquires AI-based UGC team Infinite Canvas: Again, don’t read this and think “huh, games, interesting but doesn’t apply to me” - read this and wonder “gosh, can I imagine any possible scenario in which tailored, dynamic experiences could be useful in my world?” > > “Infinite Canvas has created AI-powered experiences across UGC platforms like Roblox and Discord, reaching a peak of 19 million monthly active users. This acquisition positions Pocket Worlds to beef up its creator tools, enabling more tailored, dynamic experiences for its diverse Highrise community.”
Anthropic builds RAG directly into Claude models with new Citations API: “On Thursday, Anthropic announced Citations, a new API feature that helps Claude models avoid confabulations (also called hallucinations) by linking their responses directly to source documents. The feature lets developers add documents to Claude's context window, enabling the model to automatically cite specific passages it uses to generate answers.”
How to organise your teams - The dangerous path to Explorers, Villagers and Town Planners: There is so much good stuff in this article - these three pieces are just some of my favs.
Explorers are brilliant people. They are able to explore never-before-discovered concepts, the uncharted land. They show you wonder but they fail a lot. Half the time the thing doesn’t work properly. You wouldn’t trust what they build. They create ‘crazy’ ideas. Their type of innovation is what we call core research. They make future success possible. Most of the time we look at them and go “what?”, “I don’t understand?” and “is that magic?”. In the past, we often burnt them at the stake. They built the first ever electric source (the Parthian Battery, 400AD) and the first ever digital computer (Z3, 1943).
Villagers are brilliant people. They can turn the half baked thing into something useful for a larger audience. They build trust. They build understanding. They make the possible future actually happen. They turn the prototype into a product, make it manufacturable, listen to customers and turn it profitable. Their innovation is what we tend to think of as applied research and differentiation. They built the first ever computer products (e.g. IBM 650 and onwards), the first generators (Hippolyte Pixii, Siemens Generators).
Town Planners are brilliant people. They are able to take something and industrialise it taking advantage of economies of scale. This requires immense skill. You trust what they build. They find ways to make things faster, better, smaller, more efficient, more economic and good enough. They build the services that explorers build upon. Their type of innovation is industrial research. They take something that exists and turn it into a commodity or a utility (e.g. with Electricity, then Edison, Tesla and Westinghouse). They are the industrial giants we depend upon.
Are Cell Phones Really Destroying Kids’ Mental Health?: I’ve always had a problem with Jonathan Haidt’s hyperbolic work in the Anxious Generation. This article clarifies that.
Meet VideoRAG: A Retrieval-Augmented Generation (RAG) Framework Leveraging Video Content for Enhanced Query Responses: This is a key for leveraging libraries of video content “In a nutshell, VideoRAG represents a big step forward in retrieval-augmented generation systems because it leverages video content to enhance response quality.”