At a roundtable in Panama, futurist David Wood, SingularityNET’s Ben Goertzel, and OpenAI CEO Sam Altman examine the progress of transformer-derived AI systems toward artificial general intelligence. They compare benchmark achievements like ARC-AGI, emerging autonomous platforms such as Manus, and stress proactive governance to mitigate existential and ethical risks.

Key points

• Transformers underpin current AI, with GPT and Claude models leveraging self-attention to process vast datasets and generate human-like language outputs.
• OpenAI’s ARC-AGI benchmark scores (75.7% for o3 vs. 5% for GPT-4o) signal rapid improvements in AI reasoning, marking a leap toward generalized intelligence.
• Emerging compound AI systems like China’s Manus platform integrate multiple specialized models for autonomous task execution, foreshadowing multi-agent architectures in AGI development.

Why it matters: Understanding the trajectory toward artificial general intelligence is essential to shape policy, ensure safe development, and prevent irreversible societal impacts from unsupervised AI autonomy.

At Baylor College of Medicine, Hecker and colleagues demonstrate that psilocybin’s active metabolite, psilocin, extends human lung and skin cell lifespans by up to 57%, preserves telomere length, and reduces oxidative stress. In parallel, monthly psilocybin dosing in older mice improves survival rates to 80%, reverses hair loss, and restores hair color, highlighting its potential to modulate key aging hallmarks through telomere maintenance and stress resilience mechanisms.

Key points

• Psilocin extends human lung and skin cell lifespan by up to 57%, maintaining telomere length and reducing oxidative stress.
• Monthly psilocybin administration in aged mice increases survival from 50% to 80%, promotes fur regrowth, and reverses hair graying.
• Study highlights the role of telomere stabilization and Sirt1-mediated stress resilience as mechanisms underpinning psilocybin’s anti-aging effects.

Why it matters: This study reveals that a psychedelic compound can directly target aging biomarkers, potentially opening novel therapeutic strategies for age-related diseases.

A team led by geneticists at the University of Bath and the National University of Mexico analyzes genome data from 46 mammalian species, uncovering that gene duplication events in immune-related families correlate with both increased maximum lifespan and larger brain volumes. By comparing gene family sizes across species with high-quality genome assemblies, they identify expansions in DNA repair and inflammatory gene clusters that may underlie enhanced tissue maintenance and neural development in long-lived mammals.

Key points

• Comparative genomics of 46 mammalian species with ≥80% genome completeness identifies correlations between gene family size and lifespan metrics.
• Expanded gene families include DNA repair genes (e.g., p53 pathway components) and proinflammatory cytokines (notably interleukin-6), implicating dosage effects.
• Phylogenetic regression analysis links gene duplication rates in immunity clusters with maximum lifespan and brain size, controlling for genome quality and evolutionary relatedness.

Why it matters: Linking immune gene duplications to longevity and brain size reveals potential targets for anti-aging therapies and insights into neural maintenance.