Blog by Evelien Verschroeven: “As AI continues to evolve, it is essential to move beyond focusing solely on individual behavior changes. The individual input — whether through behavior, data, or critical content — remains important. New data and fresh perspectives are necessary for AI to continue learning, growing, and improving its relevance. However, as we head into what some are calling the golden years of AI, it’s critical to acknowledge a potential challenge: within five years, it is predicted that 50% of AI-generated content will be based on AI-created material, creating a risk of inbreeding where AI learns from itself, rather than from the diversity of human experience and knowledge.
Platforms like Google’s AI for Social Good and Unanimous AI’s Swarm play pivotal roles in breaking this cycle. By encouraging the aggregation of real-world data, they add new content that can influence and shape AI’s evolution. While they focus on individual data contributions, they also help keep AI systems grounded in real-world scenarios, ensuring that the content remains critical and diverse.
However, human oversight is key. AI systems, even with the best intentions, are still learning from patterns that humans provide. It’s essential that AI continues to receive diverse human input, so that its understanding remains grounded in real-world perspectives. AI should be continuously checked and guided by human creativity, critical thinking, and social contexts, to ensure that it doesn’t become isolated or too self-referential.
As we continue advancing AI, it is crucial to embrace relational cognition and collective intelligence. This approach will allow AI to address both individual and collective needs, enhancing not only personal development but also strengthening social bonds and fostering more resilient, adaptive communities…(More)”.