FAQ
Frequently asked questions about AGI, Hyperon, and this manual
Frequently Asked Questions
Q: Is Hyperon the only way to build AGI?
No. There are many other approaches, including pure reinforcement learning (OpenAI/Google DeepMind) and different cognitive architectures (SOAR/ACT-R). Hyperon is simply our preferred framework due to its focus on neuro-symbolic integration and scalability.
Q: How much math do I need to know?
To use the frameworks, a basic understanding of logic and probability is enough. To build and improve them, you will need deep knowledge of Linear Algebra, Information Theory, and Category Theory.
Q: Can I run Hyperon on my laptop?
Yes. The Hyperon Core and the MeTTa interpreter are designed to be efficient. However, for massive-scale experiments (billions of atoms), you will need a distributed cluster using the DAS.
Q: Is MeTTa just another Lisp?
MeTTa takes inspiration from Lisp (S-expressions) and Prolog (logic matching), but it is uniquely designed for pattern-rewriting on hypergraphs with native truth values. This makes it fundamentally different from standard Lisps.
Q: When will we reach AGI?
This is the most debated question in the field. Estimates range from 2029 to "never." Our focus is on the engineering path rather than guessing the date.
Next: References and Papers