The massive arrival of AI in businesses, with the underlying promise of replacing individuals, is reviving discussions on collective intelligence. After all, what is the point of mobilizing collective intelligence when a machine can produce a synthesis, an action plan or an idea in a few seconds?
With the fantasy of a pure and simple replacement of the human collective by artificial agents, many fears (re)emerge (AI and jobs: why I don’t believe in the “great replacement” of humans by machines and Let’s stop being nAIve with AI in the workplace).
But the question is poorly posed.
It is not collective intelligence that is in danger, but the way in which it is (or is not) implemented, which will have to evolve.
In short :
- Collective intelligence is based on diversity, the quality of interactions and a favorable environment, going well beyond simple collaboration.
- AI does not threaten the collective, it reveals the flaws in poor practices and invites us to rethink them.
- It can reinforce collective dynamics by structuring exchanges, reducing the cognitive load and promoting inclusion.
- When misused, it becomes a formatting tool, weakening critical thinking and impoverishing shared responsibility.
- It encourages the clarification of roles, the refinement of methods and the restoration of meaning to collective work as a choice and a lever for transformation.
What collective intelligence (really) is
Collective intelligence is not just about “working together” or collaborating. It is the ability of a collective to produce a reasoning, an idea, a decision that is richer than what each individual could have done alone.
It is based on several pillars, which are the diversity of points of view, the quality of interactions, a culture of debate and listening, an ability to build on the contributions of others and a framework that allows the emergenceof new ideas.
It is not the number that makes collective intelligence, but coordination, attention, trust and method.
AI is not a threat but a revealer.
In this context, AI asks us a question: what is the point of the collective if the machine produces better, faster, and more clearly?
Put this way, the answer is obvious and we can take our things and go play (The goal of the future is full unemployment, so we can play. (Arthur C. Clarke).).
But we can ask the question differently: what if AI forced us to raise the ambition of our collaborative practices?
And there the answer is much more interesting, even promising.
Because no, AI will not replace collective intelligence, provided that we practice the latter (which is not obvious everywhere) and that we practice it well. It will replace poor, ineffective, consensus-based collective practices, meetings where everyone listens to themselves talk or brainstorming sessions with no follow-up, and that is rather good news.
But it will not replace the effective practice of collective intelligence, quite the contrary.
AI as a catalyst for the collective
Indeed, AI can also be an accelerator, a stimulator of collaborative processes.
First of all, it makes it possible to structure a collective reflection even before the participants meet.
Then it can bring to light blind spots by introducing data or perspectives that the group would not have spontaneously thought of.
It can also help to reduce the cognitive load by taking care of the synthesis, formalization or analysis of data.
For some, it is finally a factor of inclusion insofar as it compensates for differences in language, culture or capacity for formalization between the participants
It then becomes a new agent in the ecosystem of cooperation, not to replace but to enrich.
Example: augmented governance
Let’s take a concrete case: that of a company seeking to improve its governance.
Today, it is common for decision-making bodies to struggle to deal with complex issues, anticipate impacts or mobilize the intelligence of all stakeholders. The result is short-sighted decisions, dialogue of the deaf, or stacks of PowerPoints from which nothing ever comes.
With AI, we can imagine augmented governance, where:
- Proposals are enriched by simulations or scenarios generated by AI.
- Meetings are prepared with contextual summaries in order to focus on the topics that are really important.
- Contributions are valued continuously, even outside real time (thanks to chatbots) that capture them even outside meetings and allow them to be “recycled” and put back into the loop when necessary.
- The collective memory is equipped, which reduces repetition and improves traceability.
Here, human power is not reduced, but rather given a more informed, structured and shared foundation.
AI is neutral, but its use is not
I know that this is a controversial point of view, but I have always been convinced that technology is neutral (Is technology really evil?). On the other hand, humans can use it for better or for worse.
And indeed, I have no doubt that some businesses will have other intentions with regard to AI.
They will want to make it a tool for standardization, which formats responses with automatic “best practices”, an instrument of intellectual laziness, where suggestions are followed uncritically, or even a substitute for debate, on the pretext that a model “already has the answer”…
This is, moreover, the direction things can take even without any malicious intent, but if a form of laziness is allowed to set in (The Impact of Generative AI on Critical Thinking)
Then yes, it becomes a hindrance. Not only to collective intelligence, but to our collective responsibility with regard to our work and its quality.
What AI can really change
Remember: technology doesn’t solve any problems, but is merely an extension of ourselves when we try to do so (Technology Doesn’t Solve Problems).
AI therefore does not kill collective intelligence, but increases its potential or reveals its shortcomings.
But to get the most out of it, it requires us to rethink and refine our collective intelligence mechanisms.
This involves clarifying roles: what is human, what can be assisted?
It forces us to rework our methods of collaboration, which will have to favor human added value more than the simple exchange of information.
It will force us to develop our critical thinking both with regard to others and with regard to AI itself (The rise of the AI manager).
Finally, it will make us re-examine the meaning of collective work: why do something together when it is no longer a constraint but a choice?
Bottom line
Collective intelligence is not on the verge of extinction, but rather at a turning point.
It is up to us to decide whether we will sink into the illusion of automated collaboration or whether, on the contrary, we will move towards more mature, more demanding, more impactful practices. And AI will not do it for us.
What makes the collective so rich is not just its ability to produce a deliverable, but its ability to help those who participate in it grow. And no machine will do that for us.
Visual credit: Image generated by artificial intelligence via ChatGPT (OpenAI)
