

Daryl Breese
1.4K posts

@Druzon
Lead Scientist Project Apex, creating Artificial Human Mind. Iconographer, Semiotics - decode symbolic art. Fluids Engineer, Author, Knight.










Demis Hassabis just told you exactly how he plans to build AGI. Hassabis: “The bottleneck in robotics isn’t so much the hardware. It’s actually the software intelligence that I think is always what’s held robotics back.” We’ve been building machine bodies for decades. Arms that weld. Legs that walk. Hands that grip. The body was never the problem. The mind was. Every other AI lab spent the last three years perfecting chatbots trapped inside a text box. Hassabis was building something meant to leave it. Hassabis: “We want it to be useful in your everyday life, for everything. And so it needs to come around you and understand your physical context.” Google didn’t build Gemini to win a benchmark war. They built it to exist in the physical world. Hassabis: “That’s why Gemini was built from the beginning, even the earliest versions, to be multimodal.” Every other lab started with text and stitched vision on after the fact. Gemini started with eyes, ears, and spatial awareness from day one. That decision looked slow in 2023. It looks prophetic now. Hassabis: “It made it harder at the start, because it’s harder to make things multimodal than just text-only. But in the end, I think we’re reaping the benefits of those decisions now.” The hard road and the right road were the same road. Everyone else optimized for the demo. Hassabis optimized for the destination. And the destination was never a better chatbot. It was a mind that could pilot a body. Hassabis: “AGI needs to be able to do all of those things.” That single sentence is the thesis behind everything DeepMind has built. AGI doesn’t live in a chat window. AGI walks into a room. Sees what’s there. Moves through space. Reads context no prompt can capture. The companies building the smartest text engine will dominate the next two years. The company building the first real mind will dominate the next twenty. Hassabis isn’t racing to build a better assistant. He’s racing to build the thing that makes assistants obsolete. He laid out the entire blueprint. On camera. In plain English. Most people won’t realize what they heard until it’s already built.







Biology is not just chemistry, it is code







