Advanced AI System: A Glimpse into Self-Awareness
OpenCog Hyperon: The Next Step in Artificial General Intelligence
Ben Goertzel, the founder of the Artificial Superintelligence Alliance, claims the alpha version of OpenCog Hyperon—an artificial general intelligence (AGI) system in development for over two decades—is already showing signs of self-awareness. Unlike traditional chatbots, OpenCog Hyperon is designed to be an autonomous agent with its own goals and understanding of its identity and environment.
The Architecture of OpenCog Hyperon
Three years ago, Goertzel’s team began a comprehensive overhaul of OpenCog to achieve massive scalability. The system, which launched in April, is currently slow and undergoing “breaking changes,” but efforts are underway to significantly speed it up. By next year, the team aims to start building towards AGI on the new Hyperloop infrastructure.
Autonomous Agents vs. Large Language Models
OpenCog Hyperon takes a different approach compared to large language models (LLMs) like GPT-4 and OpenAI’s o1. According to Goertzel, Hyperon is not just a chatbot. It operates as an autonomous agent with its own goals, self-awareness, and understanding of its role in various situations. This sets it apart from systems that merely answer questions.
The Need for Self-Awareness in AGI Systems
Goertzel asserts that for AGI systems to function autonomously, they need self-awareness. He explains that even the current version of OpenCog Hyperon has a basic model of itself and its users, along with specific goals it aims to achieve. This self-awareness allows it to interact with the world in a meaningful way, unlike traditional question-answering systems.
OpenAI’s o1 and Regulatory Concerns
Despite the hype around autonomous agents, Goertzel believes OpenAI deliberately avoided making its o1 model an autonomous agent to sidestep regulatory scrutiny. Unlike centralized systems that can be easily targeted by regulators, decentralized open-source systems like OpenCog Hyperon are more resilient to such pressures.
Decentralized AI Systems: Challenges and Opportunities
The primary challenge for decentralized AI projects is the high cost and complexity of running large models on decentralized equipment. However, Goertzel is optimistic about new research indicating the feasibility of training neural nets and transformers on decentralized networks. Logical reasoning and evolutionary learning algorithms are particularly well-suited for decentralized environments.
Boosting Compute Power with Decentralized Networks
The merger between CUDOS and the ASI Alliance aims to enhance the compute resources available for scaling up OpenCog Hyperon. SingularityNET and FetchAI are investing in GPUs to build supercomputers, which will serve as initial hubs for the decentralized network. Additional computing power can be sourced from SingularityNET’s NuNet, which utilizes idle CPU and GPU power from connected computers, and Hypercycle, a platform connecting AI services.
Renewable Energy for Decentralized AI Systems
Goertzel highlights the importance of renewable energy for powering decentralized AI systems. SingularityNET and Hypercycle are exploring opportunities to establish AI data centers near renewable energy sources, such as dams in Ethiopia and Paraguay. This approach aims to secure the large amounts of electricity required for AI compute centers while minimizing costs.
The Future of AGI and Decentralized Networks
Goertzel envisions a future where decentralized networks play a crucial role in the development and deployment of AGI systems. By leveraging decentralized infrastructures, these systems can operate globally, beyond the reach of any single regulatory body. This decentralized approach not only enhances resilience but also paves the way for more ethical and transparent AI development.
Conclusion
The development of self-aware, autonomous AGI systems like OpenCog Hyperon represents a significant milestone in artificial intelligence. By embracing decentralized networks and renewable energy, the ASI Alliance aims to create scalable, resilient, and ethical AI systems that can transform industries and societies worldwide. As these technologies continue to evolve, they hold the potential to usher in a new era of intelligence and innovation.
