Key Takeaways:
- Open Cog Hyperon: Implement early self-awareness with an autonomous AGI system.
- development timeline: In progress since 2001, the alpha version was released in April.
- scalability: A major rebuild took place by the end of the year to improve speed and performance.
- Decentralized AI: Merged with CUDOS to increase computing resources.
- future vision: Combining centralized and distributed infrastructure for global AGI development.
Open Cog Hyperon Development began in 2001 and the first open source version was released in 2008. This system differs from existing large language models (LLMs) such as GPT-4 in that it integrates logical reasoning, evolutionary learning, and dynamic knowledge graphs. This approach allows systems to adapt and update themselves, laying the foundation for true AGI.
One of the main goals of the project is scalability. Goertzel’s team began a complete rebuild of OpenCog three years ago and has achieved massive accomplishments. scalability. The alpha version was released in April, but Goertzel points out that the system is still slow and needs optimization.
The goal is to significantly increase speed, with improvements expected by the end of the year. Once this phase is complete, the team will focus on building a fully functional AGI on the Hyperon infrastructure.
start Alliance for Artificial Superintelligence (ASI)
The Artificial Superintelligence Alliance (ASI) was formed in March, uniting projects such as SingularityNET, Ocean Protocol, and FetchAI. The alliance supports the development of AGI systems, especially in decentralized and open source environments. Recent merger with CUDOS decentralized cloud The hardware network was approved by 96% of voters, giving Goertzel more computing power to further improve OpenCog Hyperon.
OpenCog Hyperon follows a different path from models such as OpenAI’s latest o1 model. According to Goertzel, OpenAI’s decision not to convert its new models into autonomous agents may be due to regulatory concerns. He believes regulators may view autonomous AGI as too risky and potentially impose strict regulations. In contrast, Goertzel’s decentralized, open-source approach makes it more difficult for any single country to control or ban AGI development.
Advantages of DCentralized AI system
One of the key advantages of decentralized AI systems is that they operate on many machines and locations globally. This reduces the risk of centralized control and opens up broader development possibilities. However, decentralized AI faces unique challenges, especially in terms of cost and energy requirements. Training large-scale neural networks is expensive and energy-intensive, which can be a barrier to scaling. distributed system.
To solve these problems, Goertzel’s team plans to combine centralized and decentralized infrastructure. They allocated a significant portion of their budget to purchasing powerful GPUs to support their operations. The decentralized aspect will gradually expand as additional computing power is extracted from connected systems through platforms such as NuNet and Hypercycle.
Goertzel’s vision for OpenCog Hyperon is to make AGI accessible and useful to everyone. By building decentralized systems, he hopes to avoid the pitfalls of centralization and build a powerful AGI that can contribute to a future shaped by intelligent, autonomous systems.