AI & Human Co-Improvement for Safer Co-Superintelligence
This addresses the safety and acceleration challenges in AI research for humanity, proposing a shift from self-improvement to collaborative enhancement.
The paper tackles the problem of achieving superintelligence safely by advocating for co-improvement, where AI systems and human researchers collaborate to conduct AI research together, aiming to accelerate progress and ensure safety through symbiosis.
Self-improvement is a goal currently exciting the field of AI, but is fraught with danger, and may take time to fully achieve. We advocate that a more achievable and better goal for humanity is to maximize co-improvement: collaboration between human researchers and AIs to achieve co-superintelligence. That is, specifically targeting improving AI systems' ability to work with human researchers to conduct AI research together, from ideation to experimentation, in order to both accelerate AI research and to generally endow both AIs and humans with safer superintelligence through their symbiosis. Focusing on including human research improvement in the loop will both get us there faster, and more safely.