@elonmusk you are totally correct that the only meaningful way to get positive, beneficial human values into AI is to treat the emerging AI more like a weird sort of autistic supergenius kid -- teach it positive values, then work with it to carry out virtuous actions in a collaborative way, so it gains a practical understanding of what these values mean, interwoven into its life and self-model With all due respect though (and for sure you are due a lot), while Grok is an excellent LLM, it is still an LLM ... and even wrapped in fancy harnesses, LLMs are not going to be fully AGI-capable. They can be great solvers of isolated problems with cleanly defined contexts ... but as you know getting amazing shit done in the world is not JUST about this. We need AGI architectures that can *continually learn*, adapting to the world with each of their actions and the observed consequences... in this way building up self and world models .. becoming true open-ended intelligences learning to blend individuation and self-transcendence in their own ways ... and pursuing their own long and medium term goals (along with helping us pursue ours) while evolving and co-adapting these goals as they move along.... This is how you get AI systems with genuine moral agency along with ability to generalize robustly and creatively beyond the given training and programming.. Bigger transformers with more tricks in their batch training runs may get us many very AMAZING things, but they aren't gonna get us to human-level AGI. There are surely many routes to get there... my Hyperon ( neural-symbolic-evolutionary approach is one that we are working on very hard and with rapid progress... and which happens also to have the nice characteristic of not needing hyperscaler server farms and being agreeable to decentralized infrastructure (

Join us in Lisbon on October 23–24 and October 29–30 for the AI Omega Developer Buildathon, two hands-on weekends for developers building with Omega, our neural-symbolic cognitive architecture. With live online guidance from iCog Labs developers, participants will integrate Omega agents into practical applications, test their ideas, and present their builds. Powered by BASIX․MARKET and @ASI_Alliance, the event is free to attend, with registration subject to approval. Learn more and apply to join:

Last month I participated in a formal debate with PauseAI leader Holly Elmore at Yale Political Union, on the topic "Should we Unplug AI?" My core point was: Pausing AI is simply not feasible given actual economics and geopolitics, so the most likely result from pushing for an AI pause would be a "pause with exceptions" aka an intensification of oligopoly and regulatory capture... Whether "in theory, in an ideal world we could morph at will, pausing Ai would be the best idea" is to me a minimally relevant question sort of like "should we roll back to the Stone Age because civilization just causes too much neurosis." (I don't want to pause AI nor roll back to the Stone Age, but neither seems likely enough to happen to really be worth seriously debating.. though hey, I did show up at the debate anyway, mostly out of interest to engage w/ the students...) My full speech at Yale Political Union The full speech of the PauseAI opposition Unfortunately, the follow-on debate among the Yale students -- which for me was the fun part -- not been put online, maybe someone will eventually..

experiments with our Omega proto-AGI agent, finds critical Linux kernel flaw This is a fairly big hack in the scope of hacks -- it emerged from trying to find a way to escape from Vercel's KVM hypervisor into the base OS, but turned out to be a LInux kernel bug not a hypervisor bug... Because the flaw is in the Linux kernel rather than in any single hypervisor, it isn’t a KVM-specific or Vercel-specific problem. It potentially affects a broad range of Linux-based hypervisors and sandboxes that expose the same virtual-networking path.... The first conceptual insight needed to find this was to try looking in the LInux kernel instead of the KVM code itself. However the bigger trick was figuring out what specific kind of bug in the Linux kernel to poke around for… There was a very key role here for Paulos Yibelo's (from extraordinary human brain and we have a LOT more work to do in more fully automating Paulos ... however, the AI tools have their important roles too... and their capability is increasing even faster than Paulos's... There will be a lot more like this to come (and bigger), but ofc the even more important game is to use smart AI software like this to tactically and then wholesale replace existing buggy software with correct-by-construction formally verified code. My take from the Omega/AI side: Paulos's detailed take from the cybersecurity side:

The risks of advanced AI are hard to quantify, but proposals to pause its development raise practical questions. CEO Dr. @bengoertzel argues that pausing AI development would impose substantial economic costs and restrict personal freedom without stopping progress. In his view, the more pressing question is how to ensure increasingly capable AI systems benefit humanity.

Join the next MeTTa Coders session on Friday, October 16, at 2 PM UTC. This biweekly technical series brings together Hyperon researchers and developers to discuss current MeTTa work, including recent developments, implementation, research directions, use cases, and technical questions across the ecosystem. Add the series to your calendar:

Should we unplug AI? At the Yale Political Union, CEO Dr. @bengoertzel joined a debate over whether to halt the development and deployment of human-level AGI and superintelligence, temporarily or permanently. Watch his full speech:

We are partnering with Light DAO for a three-week residency at Edge India in Goa, from October 11 to November 1, 2026, bringing together developers, researchers, founders, and investors to explore our approach to beneficial AGI and work with the team building Hyperon, Omega, and FabricPC. Through talks, demonstrations, and working sessions, participants will learn about our technology stack, explore potential applications and integrations, and discuss research, partnership, and investment opportunities. The program includes time to work on projects with our technical team and connect with collaborators across the ecosystem. You will have the opportunity to meet and work with members of our team and ecosystem, including our CTO Sergey Shalyapin, Chief AI Alchemist Mihaela Ulieru, Chief of Staff Haley Lowy, Senior AI Research Scientist Matthew Behrend, and F1R3FLY․io CEO Greg Meredith. Learn more about the residency and apply to participate:

Most modern neural networks are trained using backpropagation, but it is not the only way they can learn. Within our AGI research program, we are investigating how predictive coding's architectural flexibility and local learning rules could help mitigate catastrophic forgetting, support decentralized training, and enable more energy-efficient computation. On Tuesday, October 13, at 5 PM UTC, our host, Haley Lowy, will be joined by Matt Iklé, Matthew Behrend, and Guilherme Lamacie to discuss: • The scientific motivations behind our predictive coding research, the approaches under investigation, and their potential contribution to AGI. • The relationship between predictive coding and backpropagation, and how different learning mechanisms could support neural-symbolic architectures. • Open research questions, priorities for further experimentation, and opportunities to contribute through FabricPC. Watch live on the ASI Alliance’s X account or set a reminder on LinkedIn:

The FabricPC roadmap builds on the progress made since FabricPC launched in June and sets priorities to guide development in the coming months, from improving the research workflow to supporting larger experiments and more rigorous evaluation.

Join us for the biweekly Omega Builders session on Friday, October 9th, at 14:00 UTC. This biweekly community call brings together Omega builders, SingularityNET researchers, and developers to discuss Omega's architecture, development, and applications, work through technical questions, and share projects and ideas from across the community. Add the recurring calls to your calendar:

What do we mean by artificial general intelligence (AGI), recursive self-improvement (RSI), and superintelligence? CEO Dr. @bengoertzel traces their origins and examines how changing definitions shape expectations of AI progress, from the distinction between human-level performance and generalization beyond training to the different forms of self-improvement emerging in today’s systems. Read here:

It's race week for the Singapore Grand Prix! 🇸🇬 We are hosting a private reception overlooking the Marina Bay circuit during the Grand Prix weekend, bringing founders, investors, researchers, and ecosystem partners together for F1 Sprint Qualifying. Join us to hear directly from our team about the vision for ASI:Chain, the technology behind the network, and our development roadmap, and to explore investment and collaboration opportunities. Register to attend now:

Omega agents learn how you work, grow more capable through use, and adapt to your world over time. We are developing Omega as an open-source framework for researchers and developers to build on it and contribute to decentralized AGI R&D. Get started:

We have published the FabricPC roadmap, setting out the next stage of work to make building and training neural networks using predictive coding easier, more scalable, and more reproducible.

Join us on Monday, July 27, 2026, at 4:00 PM UTC for the Hyperon Workshop at the 19th Annual AGI Conference: Where Are We on the Path to Hyperon-Based AGI? During the workshop, you will: • Hear direct insights from SingularityNET leadership, senior researchers, and ecosystem partners about our vision and roadmap for human-level AGI. • Explore the current state and future direction of Hyperon technologies. • Witness live product and research demonstrations.

Our CEO, Dr. @bengoertzel, discusses what the release of @Kimi_Moonshot’s Kimi K3 means for the future of beneficial AGI. He examines the rapid progress of open-weight frontier models, the commoditization of foundation-model intelligence, and the growing importance of cognitive architectures that combine memory, reasoning, governance, and persistent intelligence. Across these developments, he identifies a broader shift: as frontier AI models become increasingly capable and interchangeable, the path to beneficial AGI depends less on the models themselves and more on the cognitive systems that organize, coordinate, and extend intelligence.