Researchers from Loughborough University, MIT, and Yale, have introduced the concept of ‘Collective AI.’ 

Documenting their ideas in a perspective paper published in Nature Machine Intelligence, researchers propose Shared Experience Lifelong Learning (ShELL) as a framework for creating decentralized AI systems composed of multiple independent agents, or “Collective AI.”

Working like a ‘hive mind,’ these individual AI units constantly learn and share knowledge throughout their lifetimes, difficult centralized monolithic architectures. 

If developed, Collective AI could mirror the talents of Star Trek’s “The Borg” and various other sci-fi concepts like “The Get” from Mass Effect or “The Replicators” from Stargate SG-1.

By enabling agents to learn from their very own experiences and the knowledge shared by others, ShELL systems can exhibit faster learning, improved performance, and greater flexibility within the face of adversity – just like biological organisms. 

Dr. Andrea Soltoggio of Loughborough University, the study’s lead researcher, described the study’s vision: “Instant knowledge sharing across a collective network of AI units able to constantly learning and adapting to latest data will enable rapid responses to novel situations, challenges, or threats.” 

Soltoggio further highlighted the potential of decentralized AI by drawing an analogy to the human immune system, where multiple components work together to mount a coordinated defense against threats. 

“It could also result in the event of disaster response robots that may quickly adapt to the conditions they’re dispatched in, or personalised medical agents that improve health outcomes by merging cutting-edge medical knowledge with patient-specific information,” Soltoggio explained. 

Several potential real-world uses are mentioned within the study:

  1. Space exploration: ShELL’s decentralized learning and adaptation capabilities might be helpful in deep space missions where communication with Earth is proscribed and autonomous systems must address unexpected challenges.
  2. Personalized medicine: ShELL could power distributed medical AI systems that continually adapt to evolving patient needs and medical knowledge, enabling more targeted and effective healthcare delivery.
  3. Cybersecurity: The collective learning and knowledge sharing of ShELL agents might be leveraged to create decentralized defensive systems that rapidly detect and disseminate details about latest threats, allowing for faster and more robust responses to cyber attacks.
  4. Disaster response: The paper suggests that ShELL systems might be used to coordinate autonomous agents in disaster scenarios, enabling more efficient and effective response efforts by leveraging the group’s collective intelligence.
  5. Multi-agent sensing: ShELL could enable the coordination of swarms of agents to construct 3D world models for tasks like search and rescue operations or anomaly detection in military reconnaissance.

Interest in decentralized AI itself is growing, as indicated by the recent resignation of Stability AI CEO Emad Mostaque who

Systems grow to be considerably more resilient when working together each collectively and independently, which we are able to observe in natural systems comparable to shoals of fish and the coordinated movements of birds and insects. 

There has been some past research in decentralized AI. For instance, a startup founded by ex-Google engineers, Sakana, recently raised $30 million for “swarm” AI that’s conceptually just like what’s proposed on this latest study. 

Building Collective AI

So how might Collective AI work? Researchers propose several potential mechanisms:

  1. Lifelong machine learning: Enables AI agents to learn multiple tasks incrementally without affected by catastrophic forgetting. Techniques include replay methods (storing and replaying previous experiences), regularization (constraining model updates to stop overwriting old knowledge), and parameter isolation (dedicating separate model components for various tasks).
  2. Federated learning: A distributed learning paradigm where multiple agents collaboratively train a model while keeping their data localized. Each agent computes model updates based on its local data and shares only these updates with others, preserving data privacy.
  3. Multi-agent systems: Study of autonomous agents interacting in a shared environment. ShELL agents operate in a decentralized manner, making decisions based on their individual goals and knowledge.
  4. Edge computing: Performing computation and data storage near the sources of knowledge, comparable to on devices or edge servers, somewhat than in centralized cloud systems. ShELL agents operate on edge devices, enabling low-latency processing and reducing communication costs.

Researchers are also mindful of the potential risks of collective AI systems, comparable to the rapid dissemination of incorrect, unsafe, or unethical knowledge between units. 

To combat that, they suggest promoting the autonomy of every AI unit inside the collective, ensuring a balance of cooperation and independence. 

Collective AI builds on recent futurist developments in AI, comparable to bio-inspired AI architectures that effectively simulate analog synaptic structures and AI models that run on real brain cells.

This article was originally published at