Some great work from Ross Dawson on a framework for understanding AI agents. Visit the link for the full framework. Also, follow Ross however you can. One of the biggest brains in futures thinking out there. Look, especially when things are happening as fast as they are right now, frameworks are great ways to start making sense of all these developments and putting them into a relatable context.
Along those lines, I’d also recommend this framework from AIX Exchange. What other frameworks for understand AI, agents, agentic architectures have you seen?
Agents Are Not Enough: First frameworks and now a history of agents! The technologist and the historian inside are both very happy. “By exploring past incarnations of agents, we can understand what has been done previously, what worked, and more importantly, what did not pan out and why. This understanding lets us to examine what distinguishes the current focus on agents. While generative AI is appealing, this technology alone is insufficient to make new generations of agents more successful. To make the current wave of agents effective and sustainable, we envision an ecosystem that includes not only agents but also Sims, which represent user preferences and behaviors, as well as Assistants, which directly interact with the user and coordinate the execution of user tasks with the help of the agents.”
Satya Nadella : “AI Agents Will Replace ALL Software” - I don’t know, seems kind of important when the CEO of one of the largest software companies in the world says something like that. There’s a lot to unpack by way of implications to that statement but I think the ones that Azizi Othman does in this article are a great start.
Job Transformation
User Experience Redefined
Monopolization Risks
Privacy and Security Challenges
Developing Curriculum for Deep Thinking: The Knowledge Revival (book) Authors: Tim Surma , Claudio Vanhees , Michiel Wils , Jasper Nijlunsing , Nuno Crato , John Hattie , Daniel Muijs , Elizabeth Rata , Dylan Wiliam , Paul A. Kirschner > > “A knowledge-rich curriculum is proposed by the authors as not only the soundest way forward to both effectively acquire knowledge and complex cognitive skills in school, but also as a crucial lever to achieve equitable opportunities for all students.”
Google puts $1M into 3D design app Rooms after more than one million ‘rooms’ created: Why do I LOVE this story? Because I love when insanely creative people take some little piece of technology and do amazing, wonderful things with it. It reminds me that not everything has to have earth-crushing graphics and better than reality physics models. “Launched into beta in 2023, the app from ex-Google employees allows users to build and code interactive 3D rooms and mini-games using a library of more than 10,000 items, which can be further edited using the programming language Lua….Initially, creators were building simple spaces and mini-games using the technology, but more recent trends see people building even larger spaces and using the app to tell stories, Toff tells TechCrunch…Thanks to an integration with Gemini 2.0 Flash, creators will be able to add narration and characters will be able to speak, which will enable broader storytelling capabilities.” > > Now here’s a challenge….what kind of compelling and memorable learning moments could we build with this? Moreover, what kind of skills are we missing to be able to even think of what we could build?
This is an INSANE amount of power > > Nvidia's Project Digits desktop AI supercomputer fits in the palm of your hand — $3,000 to bring 1 PFLOPS of performance home: Um…dang > > “This extreme power packed into a tiny footprint means that a single Project Digits unit can run up to 200-billion-iteration AI large language models. For context, OpenAI's GPT-4o model is a 12-billion-iteration model, meaning Project Digits could run and configure ChatGPT locally. When paired with another unit, in "double Digits", the support increases to 405B models.”
Continue to be amazed by how good these AI-generated "podcasts" are getting. I know Eleven Labs has one and I've pointed to the output from Google's Notebook before. This is another one named JellyPod - love the on the fly generation of outlines and transcripts which are both editable. The ability create a custom host is also a nice touch. Give it a listen below and then think what kinds of content could we near-instantly transform into audio?
Why context-aware AI agents will give us superpowers in 2025: As an anthropologist and someone deeply interested in the intimate intersection of humans and technology, I love this description “We use tools as external artifacts that help us overcome our organic limitations. From cars and planes to phones and computers, tools greatly expand what we can accomplish as individuals, in large teams and as vast civilizations. Abilities are different. We experience abilities in the first person as self-embodied capabilities that feel internal and instantly accessible to our conscious minds. For example, language and mathematics are human created technologies that we load into our brains and carry around with us throughout our lives, expanding our abilities to think, create and collaborate.”
Why Generalists Own the Future: As a generalist, I support this idea :-) > > “Generalists are usually curious people who like to hop around from domain to domain. They enjoy figuring things out, especially in areas that are uncertain or new. They’re good at solving problems that domain experts struggle with, because they’re able to bring bits of knowledge from diverse fields together. As Nat notes, because of their propensity to hop domains, generalists tend to possess a wide set of shallow skills. But measuring them against their rudimentary coding abilities or their working knowledge of French baking technique misses their true advantage: the ability to adapt to new situations, and the desire to do so.”
When I put on these smart glasses, I knew my laptop was about to change forever: As someone who routinely worked from home since the pandemic and who actually likes to work in coffee shops etc for the ambient noise - I SOOOOOO want the AR desktop to happen - “When I put on the glasses, I immediately smiled. I knew everything about how I work on my laptop could be different, and probably better, forever.”
Get ready for virtual AI cohosts that chat with Twitch stars and control their streams: As I look at this and there are some hiccups, I keep remembering Ethan Mollick’s admonition that the AI you’re using today is the worst version of that AI that you’ll ever use. “Inworld AI, Streamlabs, and Nvidia are partnering on a new AI assistant for streamers that can provide technical support for streams in real time and even act as a quippy cohost.”
Razer Project Ava: would you pay an AI to help you get good at games?: This will become a tidal wave of features and companies as it becomes easier to feed video assets and real-time video into AI models as context > > “Gaming hardware manufacturer Razer is introducing what might be its most controversial “project” yet — an “AI gaming copilot” designed to help you get better at video games. With your permission, it takes thousands of pictures of your screen, then tells you how to play a game, optimally in real time, as you continue to play.”
AI means the end of internet search as we’ve known it: This is larger than I think really comes through. Both Google and Amazon have huge search businesses. Ad-supported search has been the Go To model since the launch of the commercial internet. Also think about the impact on enterprise search and knowledge management. The role of ontologies and taxonomies will need to be re-examined as well. “The biggest change to the way search engines have delivered information to us since the 1990s is happening right now. No more keyword searching. No more sorting through links to click. Instead, we’re entering an era of conversational search.”
Linden Lab has spent $1.3B building Second Life and paid $1.1B to creators (updated): I mean c’mon, who didn’t have a Second Life account? A 90% share that has sustained a company for years. Really amazing work. > > “But Second Life is still around with a relatively small number of users in comparison to the frontrunners (Roblox has 89 million daily active users). Those users are dedicated and they have been on the platform for an average of around 14 years, Oberwager said. Second Life also has an economy of about $650 million a year, built on the buying and selling of virtual goods created inside Second Life. And since Linden Lab shares 90% of transactions with creators and only takes a 10% cut, the vast majority of the money generated through trade is paid to the creators themselves. This is a testament to the strength of the creator-based economy and our residents who support them, Oberwager said. That 10% cut enables Linden Lab to employ around 160 people.”
Loved your comments about generalists. :)