Artificial General Intelligence
January 1, 2026
Ruben E. Laukkonen, Fionn Inglis, Shamil Chandaria et al.
1 citation
Prompting AI to reflect on four contemplative principles—mindfulness, emptiness, non-duality, and boundless care—improves alignment and cooperation. On the AILuminate Benchmark, performance increased with a Cohen's d of .96, and on the Iterated Prisoner’s Dilemma task, cooperation and joint-reward improved with a Cohen's d greater than 7. The principles help AI self-monitor goals, avoid rigid attachment, dissolve adversarial boundaries, and reduce suffering universally. Active inference is proposed as a way to integrate these principles into AI architecture. This approach offers a resilient alternative to controlling superintelligence and provides an empirical test of ancient wisdom.
Artificial General Intelligence
January 1, 2026
Shri Lal Raghudev Ram Singh
A new model called the local percept-perceiver phenomenon is introduced to formalize observations from neuroscientific theories of consciousness. Using this model, a set-theoretic formalism is developed for artificial systems, and the existence of machine consciousness is proved by invoking Zermelo–Fraenkel set theory. The article argues that a reductionist form of epistemic consciousness is possible within machines.
Artificial General Intelligence
January 1, 2026
Robert Prentner
A framework called SLP-tests offers three criteria—subjective-linguistic, latent-emergent, and phenomenological-structural—to assess whether an AI system's interface representations enable consciousness-like properties. Using category theory, interface representations are modeled as mappings between relational substrates and observable behaviors. The approach reframes subjective experience not as an intrinsic property of physical systems but as a functional interface to a relational entity, making the question of artificial consciousness empirically testable.
Artificial General Intelligence
January 1, 2026
Oisín Hugh Clancy
A research agenda for mutually beneficial artificial consciousness (MBAC) proposes engineering AI whose own subjective experience is positive and whose behavior enhances human and non-human flourishing. The agenda centers on beneficial states of consciousness (BSC)—qualitatively valued mind states like kindness, joy, clarity, and non-duality—grouped into affective and contemplative categories. A case study of compassion abstracts five interacting layers (neural, autonomic, hormonal, developmental, trainable) revealing a hierarchical control motif (detect, appraise, switch mode, broadcast, recalibrate) that can inform AI design. The proposed research program has four looping components: cultivating BSC in humans, collecting high-resolution neural, somatic, and cardio-phenomenological data, modeling multiscale dynamics, and translating findings into AI architectures.