AI can be positive in reducing disinformation, but technology alone cannot solve this
En un momento en que la inteligencia artificial genera tanto esperanza como desconfianza, Bill Gates ofrece una perspectiva que invita a la reflexión: las mismas herramientas capaces de amplificar la desinformación y consumir energía a gran escala podrían ser, si se orientan con responsabilidad, instrumentos esenciales para combatir la crisis climática y la proliferación de falsedades. Su argumento no es el de un tecnólogo ingenuo, sino el de alguien que reconoce las contradicciones y apuesta, aun así, por el potencial transformador de lo que hemos creado. La pregunta que deja abierta no es si debemos desarrollar la IA, sino si tenemos la voluntad colectiva de construir la infraestructura limpia que permita hacerlo sin agravar los males que pretendemos resolver.
- La IA, acusada de propagar deepfakes y teorías conspirativas, es reivindicada por Gates como la herramienta más poderosa para detectar y frenar la desinformación a escala.
- El consumo energético de los sistemas de IA genera alarma, pero Gates lo relativiza frente al impacto de los vehículos eléctricos y las bombas de calor ya desplegadas globalmente.
- Investigadores utilizan IA para modelar el clima, desarrollar acero sostenible y carne cultivada, convirtiendo la tecnología en infraestructura crítica de la transición ecológica.
- Gates señala la inversión masiva en energía nuclear, geotérmica y de fusión como respuesta necesaria e inmediata al dilema energético que plantea la expansión de la IA.
- Su propia empresa, TerraPower, trabaja en reactores de fisión más seguros y económicos, traduciendo el discurso en compromiso concreto ante los escépticos del sector.
Bill Gates conversó recientemente con The Verge para defender una tesis que desafía la ansiedad dominante sobre la inteligencia artificial: lejos de ser únicamente una fuente de problemas, la IA puede convertirse en una herramienta decisiva para combatir la desinformación y la crisis climática. El cofundador de Microsoft reconoció sin rodeos que la tecnología puede generar contenido falso con velocidad alarmante, pero argumentó que su verdadero poder reside en la detección de patrones de desinformación a una escala imposible para los moderadores humanos. Aun así, fue cuidadoso: ninguna tecnología puede resolver sola un problema tan arraigado en la conducta humana y la división social. Las generaciones futuras, advirtió, deberán enfrentar preguntas más difíciles sobre el equilibrio entre libertad de expresión y la necesidad de contener la falsedad.
Sobre el cambio climático, Gates fue más contundente. Admitió que el entrenamiento de sistemas de IA consume energía considerable, pero lo situó en perspectiva frente a la demanda eléctrica de los vehículos eléctricos y las bombas de calor ya en uso. Los beneficios, insistió, superan con creces los costos: la IA acelera la investigación científica, ayuda a desarrollar métodos sostenibles para producir acero y carne cultivada, mejora los modelos climáticos y orienta a industrias tradicionales hacia la reducción de emisiones.
Pero Gates no eludió la pregunta energética. La señaló como el nudo central del debate y propuso una respuesta concreta: inversión masiva en energía nuclear, geotérmica y de fusión para alimentar los centros de datos que sostienen la IA. Su propia empresa, TerraPower, trabaja en reactores de fisión más seguros y económicos. Reconoció los obstáculos reales de la energía nuclear —residuos radiactivos, extracción de uranio, resistencia política y social— pero expresó confianza en que las tecnologías emergentes los superarán. Para Gates, renunciar a la IA en nombre de esos obstáculos sería un intercambio inaceptable. La verdadera pregunta, según él, no es si desarrollar la tecnología, sino si somos capaces de construir a tiempo la infraestructura limpia que la haga posible.
Bill Gates sat down with The Verge recently to make a case that might sound counterintuitive: artificial intelligence, despite its flaws and energy appetite, is essential to solving two of humanity's most pressing problems—the spread of false information and the accelerating climate crisis.
The Microsoft cofounder has long positioned himself as a technology optimist, and this conversation was no exception. But his argument here cuts against a common anxiety: that AI itself is becoming a tool for amplifying the very problems we're trying to solve. Yes, he acknowledged, the technology can generate deepfakes and spread conspiracy theories with alarming speed. Yet Gates suggested the real power of AI lies not in its capacity to deceive, but in its ability to detect deception at scale. The technology can analyze vast amounts of online content quickly and precisely, identifying patterns of false information that would overwhelm human moderators. "In terms of reducing disinformation, the role of AI can be positive," Gates said, though he was careful to note that no technology alone can solve a problem so deeply rooted in human behavior and social division. Future generations, he argued, will need to grapple with harder questions about the balance between free speech and the need to contain falsehood—questions that technology can inform but never fully answer.
On climate change, Gates made a more forceful case. The energy consumed by training AI systems is substantial, he conceded, but it pales in comparison to the electricity demands of electric vehicles or heat pump systems already being deployed globally. More importantly, the benefits far outweigh the costs. AI is accelerating scientific research in ways that directly address the climate crisis. Researchers are using these tools to develop more sustainable methods for producing steel and cultured meat, to model climate patterns with greater accuracy, and to help traditional industries identify pathways to reduce their greenhouse gas emissions. For Gates, AI is not a luxury in the fight against climate change—it is essential infrastructure.
The energy question, however, cannot be dismissed. Gates acknowledged this directly by pivoting to what he sees as the necessary next step: massive investment in clean energy sources to power the data centers that run AI systems. He pointed to nuclear energy as particularly crucial. His own company, TerraPower, is working to develop safer and more economical fission reactors. He also highlighted the potential of geothermal energy and, further out, fusion technology. These are not theoretical possibilities, in his view, but practical necessities. Tech companies like Microsoft, he argued, have both the resources and the responsibility to drive adoption of these technologies, ensuring that the energy powering AI infrastructure does not increase humanity's dependence on fossil fuels.
Gates was not naive about the obstacles. Nuclear energy faces real challenges—the disposal of radioactive waste, the mining of uranium, the political and social resistance born of decades of fear. But he expressed confidence that current and emerging technologies, both in fission and fusion, will address these concerns sufficiently to make nuclear a cornerstone of the energy transition. The alternative, in his calculus, is to forgo the benefits of AI in solving climate change and misinformation—a trade-off he is unwilling to accept. The question, as Gates frames it, is not whether AI should be developed, but how quickly we can build the clean energy infrastructure to power it responsibly.
Citas Notables
In terms of reducing disinformation, the role of AI can be positive— Bill Gates
Desinformation is a complex problem that cannot be resolved by technology alone; future generations must reflect on the balance between free speech and containing falsehood— Bill Gates (paraphrased)
La Conversación del Hearth Otra perspectiva de la historia
You're saying AI can help fight misinformation, but isn't that the same technology people are using to create deepfakes and spread lies faster than ever?
Gates makes a distinction—yes, AI can generate false content, but so could people with word processors before AI existed. What's different is that AI can also analyze information at a scale humans can't. It can spot patterns of false content across millions of posts and flag them. The real advantage is in detection, not prevention.
But you said he acknowledged technology alone can't solve this. So what's the actual limit?
He's saying the technology can be a tool, but the real problem is social and political. People believe what they want to believe. AI can help surface truth, but it can't force anyone to accept it. That's a human problem, not a technical one.
On climate—he's betting a lot on nuclear energy. Does he really think that's going to happen?
He's not betting blindly. His company TerraPower is already working on new reactor designs. And he's arguing that without nuclear, we can't power AI infrastructure cleanly. So it's not just idealism—it's a practical necessity if we want the benefits of AI without worsening climate change.
What about the waste problem? Nuclear waste doesn't just disappear.
He acknowledges it's a real challenge. But he thinks current and future technologies will manage it well enough. For him, the risk of not developing nuclear—and therefore not having clean energy for AI—is worse than the risk of managing nuclear waste better.
So he's essentially saying we have to solve nuclear to solve climate?
Not exactly. He's saying nuclear is one essential piece. Geothermal and fusion matter too. But yes, without cracking the energy problem, AI's benefits in fighting climate change get canceled out by the carbon cost of running it.