Posts

Researchers analysing 1,245 natural vocal interactions found that wild common marmosets use flexible turn-taking, matched call-response sequences, brief pauses, and very little overlap across both within-group and intergroup exchanges.

Researchers combined more than 16,000 soil samples with predictive models to map the global density of arbuscular mycorrhizal fungal networks.

Long before the dawn of modern parenting, animals laid eggs and moved on, leaving their progeny to fend for themselves. Now, a study published in Nature uncovers one of the elegant ways that evolution transformed neglect into nurture.

(See the original demos announcement at the link). We have added more algorithms to the website to try out and tweak.

Pentest Swarm AI uses a shared blackboard, pheromone-weighted findings, and decentralized agent triggers to let attack paths emerge instead of following a fixed pipeline.

This funding call frames AI safety as a population-level problem involving collusion, cascading failures, emergent communication, and collective capabilities. This closes 9-August-2026.

Palladyne AI says its SwarmOS software will be tested with U.S. Army warfighters to coordinate heterogeneous unmanned systems without relying on centralized infrastructure.

INT21’s PTX Kernel Factory uses specialized agent swarms to generate, test, benchmark, and improve low-level GPU kernels across generations. This is a practical software example of evolutionary adaptation: many agents explore candidate solutions, retain evidence, and feed successful traces into the next cycle.

Google DeepMind’s roadmap treats future AI agents as operational actors that may need containment, monitoring, and layered controls, borrowing ideas from cybersecurity and safety engineering.

DARPA’s new DICE (Decentralized Artificial Intelligence through Controlled Emergence) program may be one of the strongest real-world signals yet that biological coordination principles are moving into operational AI architectures.