The conversation about artificial intelligence often focuses on new models, benchmarks, or impressive demos.
But if you step back a bit, deeper signals start to appear.
It’s not just more powerful software. We’re seeing changes that affect how we think, how technological infrastructure is built, and how the relationship between humans and machines might evolve.
These are five signals that are emerging now — and that could define the next decade.
1️⃣ What if an AI were already conscious… and we simply don’t know how to detect it?
For years the idea of a conscious artificial intelligence belonged more to science fiction than to scientific debate.
But that conversation has started to change.
Anthropic CEO Dario Amodei has noted in interviews that we cannot completely rule out that advanced AI systems might develop some form of inner experience.
The problem is we don’t know how to detect it.
We don’t even fully understand human consciousness.
Neuroscience still debates how subjective experience arises from approximately 86 billion neurons in the human brain. If we don’t know exactly how it emerges in us, identifying something similar in a machine becomes even more complex.
Meanwhile, current AI systems keep growing rapidly. Models developed by OpenAI, Google, and Anthropic operate with hundreds of billions of parameters and are trained using enormous amounts of data and computing capacity.
The problem is that these systems largely function as black boxes. Even the engineers who build them often cannot explain exactly how they arrive at certain answers.
This opens an unexpected ethical question.
Today AI models can run on thousands of servers, be copied, shut down, and restarted at will. But if someday some form of inner experience existed —even if minimal— the situation would be completely different.
We would be talking about creating and shutting down possible intelligent entities without knowing it.
So far there is no scientific evidence of consciousness in current models. Most researchers believe they remain extremely complex statistical systems.
But the fact that this discussion is taking place in labs and tech companies shows something important:
For the first time in history we are building machines complex enough that the question of their consciousness no longer seems absurd.
Source: Machines of Loving Grace — Anthropic
2️⃣ The world’s new strategic resource: chips for artificial intelligence
For much of the twentieth century the strategic resource was oil.
In the twenty-first it could be something much smaller: advanced computing chips.
AI models require enormous amounts of computing capacity to train. Today much of that infrastructure depends on hardware produced by NVIDIA.
It is estimated that more than 80% of global AI model training uses NVIDIA GPUs.
Training advanced models can require tens of thousands of GPUs working simultaneously for weeks or months. Some clusters used by tech companies exceed 20,000 GPUs running in parallel.
This has turned AI chips into a geopolitical asset.
The United States has restricted the export of advanced chips to China to limit its ability to develop competitive models. At the same time, China is investing billions in its own semiconductor industry.
According to McKinsey & Company estimates, the global market for AI chips could exceed 400 billion dollars annually before 2030.
The AI race is no longer only about algorithms.
It is also about who controls the physical infrastructure that makes training them possible.
Source: The economic potential of generative AI — McKinsey
3️⃣ The energy problem of artificial intelligence
As models grow larger, so does their energy consumption.
The data centers that train and run AI models require enormous amounts of electricity.
According to International Energy Agency estimates, data center energy consumption could double before 2030, driven largely by artificial intelligence.
To put it in perspective:
- some AI data centers consume as much electricity as small cities
- training an advanced model can require gigawatt-hours of energy
This is pushing major tech companies to invest directly in energy infrastructure.
For example:
- Microsoft and Google are exploring the use of modular nuclear reactors to power future data centers.
- Amazon has become one of the world’s largest corporate buyers of renewable energy.
The paradox is interesting.
The most advanced technology on the planet increasingly depends on very traditional physical infrastructure: electricity, cables, and giant buildings full of servers.
Source: Electricity 2024 — International Energy Agency
4️⃣ We’re starting to delegate part of our thinking to AI
Beyond infrastructure, the deepest change may be human.
More and more people are using artificial intelligence for cognitive tasks:
- writing documents
- summarizing information
- analyzing data
- programming.
Research from Stanford University and the Massachusetts Institute of Technology has shown that AI tools can increase productivity on knowledge tasks by 20% to 40%.
This is generating a phenomenon known as cognitive offloading: delegating mental processes to external tools.
For centuries we did this with calculators, books, or search engines.
The difference is that now we don’t just search for information.
We can ask a system to analyze, synthesize, and propose complete solutions.
The question that is starting to arise is inevitable:
Will artificial intelligence expand our cognitive capacities… or will we start to depend on it too much?
Source: AI Index Report — Stanford HAI
5️⃣ What the predictions say for the next decade
If anything is clear in the tech industry, it’s that artificial intelligence is still in its early stages.
Many of the tools we use at scale today —such as code assistants or copilots— appeared only in the last few years.
But projections for the next decade suggest that the impact could be much larger.
According to McKinsey Global Institute estimates, up to 30% of current working hours could be automated before 2030.
That doesn’t necessarily mean entire jobs will disappear, but it does mean many tasks within them could change radically.
At the same time, the global artificial intelligence market could exceed 1.8 trillion dollars before the end of the decade, according to Statista data.
But the most interesting change isn’t just market size; it’s how AI will begin to integrate deeply into organizations.
Companies like OpenAI, Anthropic, and Google are already developing AI agents capable of executing full tasks, not just answering questions.
This could redefine how many teams work.
- An AI-assisted engineer can review thousands of lines of code in seconds.
- An analyst can explore datasets that previously required days of work.
- A small team can produce what used to require entire departments.
The deepest change won’t be only technological.
It will be how humans and intelligent systems start to work together.
Source: Artificial Intelligence — Statista
A final reflection
Artificial intelligence is often presented as a technological race.
But it’s becoming clearer that it’s something deeper.
We’re building systems that don’t only transform industries or software.
They’re also starting to influence how we think, how we learn, and how we make decisions.
Perhaps the most important question is no longer how intelligent the machines can become.
The question might be another:
How does humanity change when it starts to coexist with intelligences created by itself?
✍️ Claudio from ViaMind Dare to imagine, create and transform.
Also available in Spanish: Tendencias IA & Tecnología – Semana del 16 Mar 2026.
FEATURED POSTS
ABOUT ME
ViaMind Ecosystem
- ViaMind — The brand hub for ideas, technology and strategy.
- NeuraPRO — Practical software for managing real operations.
- Consulting — Advisory for complex AI and telecom decisions.