top of page

AI and Collective Intelligence

The recent progress in Artificial Intelligence provides both challenges and opportunities for scientific collective intelligence. Among the most notable examples of AI progress are ChatGPT and other Large Language Models, which have shown unexpected capabilities (Wei 2022, Mitchell 2023), and AlphaFold, which can predict the 3D shape of proteins from their genetic sequence with unprecedented accuracy (Jumper 2021).



AI poses specific challenges for science. There are many reports of errors in statements from ChatGPT and from other AI systems. The types of errors and blind spots seem different from those more common in humans.

These AI systems consist of neural networks with billions to trillions of parameters (Mitchell, 2023) and it is therefore not possible to provide a simple explanation for their outputs. This makes it difficult to evaluate any generalization to a problem different from the dataset on which the AI system has been trained.

An AI system with a good track record might suggest an experimental approach that requires large amounts of resources and we might not be able to provide an independent assessment.



We can consider AI also as an opportunity for science and collective intelligence.

1) There is wide support for the view that human intelligence evolved in response to intellectual challenges, posed by culture and by social interactions. Recently, the pace of societal change has outstripped biological evolution, and has manifested itself as non-biological cultural evolution. AI systems should be viewed as an opportunity that will stimulate the continuing cultural evolution of our collective intelligence.

There are several examples of major challenges that have promoted collective scientific efforts. Among these are:

  • The WW2 effort at Bletchley Park, where Alan Turing played a key role; this effort led to the first purely electronic digital computers, responding to cryptography advances by Germany.

  • The Manhattan Project, prompted by the discovery of nuclear fission by German scientists and leading to advances in nuclear physics, for both military and peaceful applications.

  • The creation of NASA that led to the Moon landing, sparked by the Sputnik launch and the space race with the Soviet Union.


2) AI, with its variety of ways of organizing vast amounts of information, can make large scale discussions possible by finding new ways to connect ideas. This could be an iterative process in which human scientists can decide which avenues to pursue and provide novel integrated contributions.


3) Collective intellectual efforts can be promoted by incentives, as discussed in another section. The expectation that new forms of scientific collective intelligence are likely to emerge implies that the first participants will be recognized for their pioneering efforts. This would be a self-reinforcing prediction, able to create its own incentives.


Choice of problem

The scientific problems most suitable for an approach based on AI and collective intelligence are those that are broadly relevant for society, too complex for an individual human mind or for a small-team approach and where large datasets and new technologies are available. The field of cell-cell communication has these properties. It was chosen after discussions among hundreds of scientists, starting from an initial focus on the intercellular communication of one cell type, endothelial cells.


Plan of discussion

Contributions might consist of a paragraph and up to five references. They will be connected with other ideas, initially using human suggestions and later, when required by the scale of the effort, also using AI. Other formats can be used, and the contributions can be sent to our contact email or be entered directly in the dedicated Forum section. Contributions will be saved on this website and on the Internet Archive (, to assist in providing recognition for the original proponents of novel ideas.

In the AI and Collective Intelligence section of the Forum we will discuss ideas generally applicable to AI and scientific collective intelligence, while in the rest of the website we will focus on the application of these approaches to cell-cell communication.



-Jumper, J., et al., 2021. Highly accurate protein structure prediction with AlphaFold. Nature, 596(7873), pp.583-589.

-Mitchell, M. and Krakauer, D.C., 2023. The debate over understanding in AI’s large language models. Proceedings of the National Academy of Sciences, 120(13), p.e2215907120.

-Wei J et al. 2022. Emergent abilities of large language models. arXiv preprint arXiv:2206.07682.

bottom of page