Humanity is found on the threshold of an era of deep transformation, where digital superintelligence (AGIS) seems to be at your fingertips. Sam Altman, co -founder and CEO of OpenAihe does not hesitate to affirm in his most recent publication of his personal blog “The Gentle Singularity” that “we are beyond the event horizon; the takeoff has begun.” This statement, far from being a futuristic fantasy, is based on the reality of systems that already exceed human capacities in multiple aspects, amplifying the productivity of those who use them.
This progress promises a radically better future than the present, driven by unprecedented acceleration in science and a massive increase in productivity. In this 2025 we have already seen the arrival of agents capable of performing real cognitive work, transforming tasks such as writing code or Automation of administrative backoffice tasks.
Altman projects that by 2026 systems capable of discovering novel knowledge could arise, and in 2027, robots that perform tasks in the real world. According to his gaze, the abundance of intelligence and energy for the 2030s, historically the fundamental limitations of human progress, would theoretically allow us to reach anything with good governance. Wonders will become everyday; The amazing of today, as a well written paragraph by AI, will become the expectation of a complete novel or The development of new medicinal treatments They heal all known diseases.
However, this vertiginous technological acceleration brings with it a series of inherent risks that, if they are not addressed with consciousness and proactivity, could undermine the same foundations of society. Altman himself acknowledges that, although a small new capacity for AI can generate a huge positive impact, “a small misalignment multiplied by hundreds of millions of people can cause a great negative impact.” This is the heart of the paradox we face: how to reap The immense benefits of AI without succumbing to its latent hazards.
One of the most pressing risks is the proliferation of falsification in all its forms. Altman warns about the possibility of “that machines flood our information channels with propaganda and falsehood.” The capacity of generative artificial intelligence to create Deepfakes —Refvavisual countentes falsified with surprising accuracy – raises a serious challenge to our perception of reality. Images, videos or manipulated audios can ruin individual reputations, undermine confidence in companies and, in the political sphere, distort public opinion and even influence elections. The ease with which false content can be generated and distributed, especially in a world where people have lost energy to validate sources, erodes collective trust in media and institutions, weakening social fabric. Constant exposure to misinformation can lead to generalized cynicism and social fragmentation.
Another critical challenge is the impact on the labor market. Altman mentions that “there will be very difficult parts such as the disappearance of whole jobs.” Although history shows that humanity adapts, and that new jobs will be created, the magnitude of this transformation is undeniable. Automation is not a new phenomenon, but AI accelerates this process, mainly affecting average, production and administrative stalls. While it is estimated that Automation will displace millions of jobsthe same number or more could be created, but these will be new, different. This demands a massive reconversion of the workforce and a preparation of the new generations for roles that will demand different skills, such as critical thinking, emotional intelligence and flexibility. Companies that do not invest in the training of their employees in AI and process design will be at a disadvantage.
Technological dependence is perhaps the most subtle and worrying risk. Altman himself experienced him when Openai had a fall: “When we needed it, I was not (…) I had forgotten what it was like to do. And it really made me think, we will be dependent on these systems, more and more.” This delegation of tasks and processes to automatic systems, if consciously managed, can lead to an atrophy of essential human skills. When it is blindly trusted in AI to answer questions, generate texts or solve problems, there is a risk of losing the ability to investigate, synthesize, analyze and think autonomously. This has been described as a “brain rot” (Brain Rot), where excessive consumption of trivial content or technological dependence in basic cognitive tasks can lead to a deterioration of essential cognitive functions, such as concentration and deep reflection. Opting for the least effort offered by AI without an active commitment, instead of intellectual effort, can seriously limit professional and personal growth.
Society must also face inherent biases in AI models. Altman points out the need to “solve the alignment problemwhich means that we can guarantee in a robust way that the AI systems learn and act towards what we collectively want in the long term. “A clear example of misalignment is the feeds of social networks, which exploit short -term preferences over long -term.
If the data with which the AI is trained contain biases or prejudices, the AI will replicate and amplify them. This can lead to discriminatory decisions in critical areas such as hiring or approval of loans, if algorithms are not carefully designed and audited to promote equity and inclusion. The emergence of models such as Deepseekwhich achieved high performance with much less resources than its competitors, generated a new technological rivalry and forced giants as OpenAi to reduce prices, demonstrating the volatility of the sector. However, Depseek has also been “indoctrinated” with biases of the Chinese government, which underlines the importance of human supervision in critical uses to avoid moral, ethical, monetary or legal costs.
Faced with these challenges, Altman’s optimism remains firm, but anchored in action and collaboration. No doubt we all trust the human ability to adapt. The story, according to him, shows that “we will find new things to do and new things to desire, and we will quickly assimilate new tools.” The key lies in how society decides to manage this transition. It is essential to distribute widely the access to superintelligence given its economic implications, and ensure that it is cheap, widely available, and not too concentrated in a person, company or country.
To navigate this era, it is imperative to adopt a conscious adoption approach to technology. This means deliberately evaluating the impact of each new tool, wondering if it really provides value and if it strengthens our human abilities instead of weakening them. It is not about resisting progress, but about being selective and critical, prioritizing mental health and personal relationships on mere convenience. It is essential to limit the passive use of technology and reserve moments for programmed disconnection, promoting continuous learning and challenging the use of our skills to avoid the blind delegation of creative and analytical processes.
The continuous development of human skills is our greatest competitive advantage. While the machines assume repetitive tasks, creativity, empathy, critical thinking and the ability to solve complex problems will continue to be irreplaceable. Empathy allows us to build solid relationships and understand the emotional nuances that no algorithm can replicate. Effective communication, which goes beyond words and includes tone and body language, is essential to connect with others and lead with humanity. Creativity, the engine of innovation, cannot be completely replicated by machines, since it arises from personal experience, emotions and a unique perspective in the world. Promote these skills through self-reflection, continuous training, mentoring and consistent practice is crucial.
Critical and analytical thinking, the ability to break down complex information and question the sources, becomes more vital than ever in a saturated data environment. Instead of blindly accepting the information generated by AI, it is mandatory to verify it, contrast sources and analyze coherence. Language models, based on historical data, have limitations to anticipate black swans or unpredictable events, which underlines the irreplaceability of human judgment and intuition. As Altman points out, “society is resilient, creative and adapts quickly”. If we can take advantage of collective will and wisdom, we will maximize the benefits and minimize the risks, giving people a lot of freedom within broad limits that society must define.
Ultimately, the future of AI and its impact on humanity is not a default destination, but a collective construction. This is the call to intelligent and ethical adoption. The key is not to prohibit or condition the use of AI, but to learn to live with it intelligently and ethically. By committing ourselves to the development of our human skills, cultivating curiosity, critical thinking and adaptability, we can ensure that technology is a tool at the service of our evolution, not a substitute for our essence. True progress will be measured in how these advances enrich human experience and not only in terms of efficiency. We can climb fluidly, exponentially and without incident through superintelligence, only if we do it with consciousness and purpose.
Eduardo Laens is CEO of Varegos and university and secondary teacher specialized in AI.
You may also interest you
By Eduardo Laens
Image gallery
