A line of work is beginning to take shape within the AI ecosystem that moves beyond the most media-friendly version of artificial intelligence, the one that reduces everything to chatbots, copilots and promises of productivity, and instead turns to a far more ambitious approach. What is at issue here is the attempt to build systems capable of reasoning under conditions of high uncertainty, where poor judgement does not merely create friction but can also cause real harm.
New frontier
While much of the public debate remains trapped between fantasies of mass automation and apocalyptic dread, this strategy is operating on a different plane: how to introduce forms of complex reasoning, grounded in a bioneurocognitive approach, into military and civilian systems that already exist or are still to be designed. The shift matters because it moves the discussion away from tools and towards the architecture of intelligence, defining a field of its own around complex reasoning, cognition and the design of intelligent systems. The frontier is no longer prediction so much as reasoning; no longer the generation of plausible outputs, but the ability to hold context, weigh competing hypotheses, adapt and justify decisions under pressure. We are dealing with a subtler and more demanding level of AI.
Behind this approach stands Luis Martín, a researcher and designer of advanced AI systems specialising in complex reasoning architectures, and the author of the expression “bioneurocognitive AI”, a semantic innovation that condenses this emerging technological perspective and is closely associated with his own research and design ecosystem. It is not yet a settled category within technological jargon, but rather an authorial formulation that can be situated within several active strands of the scientific literature, from biologically inspired cognitive architectures to neuro-symbolic AI and neurocognitively inspired reasoning systems. That is precisely what makes it interesting: it is not only an attempt to develop systems, but also to name a field of its own, bound up with complex reasoning, cognition and the architecture of intelligent systems. It is, in that sense, a genuine conceptual innovation with emerging applications.
Three converging fields
So what exactly are we talking about? Martín’s starting point is that AI is not one thing. It is, at the very least, the convergence of three fields: machine intelligence as it currently dominates public discussion; the enhancement of human cognitive performance; and, above all, dual intelligent ecosystems, in which humans and artificial systems work together continuously, with distributed functions, different tempos of decision-making and explicit rules of governance. Put differently, the ambition is no longer simply to automate tasks, but to decide how perception, interpretation, judgement and action are to be distributed between people and machines.
What is at stake, then, is not merely a family of technologies, but a new way of conceiving intelligent organisations: universities, public administrations, defence systems, research environments or business structures designed no longer around isolated tools, but around cooperation between human and machine intelligence. It is a perspective that opens up a great deal.
This article is republished from Futuribles. Here's the original article in Spanish and English.