The ARC Prize, a $1M competition designed to challenge advancements toward Artificial General Intelligence (AGI), has seen significant progress with its 2024 winners implementing innovative reasoning approaches. This shift marks a departure from reliance on pure deep learning frameworks, as demonstrated by successful models like ARChitects. The community's response to tackling the ARC-AGI benchmark shows evolving contributions beyond traditional methods. Noteworthy is the open-source availability of all winners' code and papers, fostering collaboration and further experimentation in the field. Additionally, concerns about task susceptibility to brute force solutions and the effectiveness of core knowledge priors have been raised, highlighting areas for refinement in future competitions (ARC-AGI-2). Overall, the ARC experiment validates the notion that new paradigms are necessary for AGI development, pushing the boundaries of current methodologies.