Inteligencia artificial: cómo entenderla sin miedo | José Ignacio Latorre, físico cuántico

Summary

José Ignacio Latorre argues that artificial intelligence and quantum computing represent major technological shifts. He believes they should be met with understanding, public discussion, and regulation—not fear or uncritical enthusiasm.

How AI developed

Latorre describes AI as an effort to reproduce aspects of intelligent reasoning through algorithms. He traces its roots to early computing and information theory, followed by neural networks, deep learning, and, more recently, Transformer models. These models process language in context and underpin systems such as GPT-style chatbots. Progress has depended partly on larger networks and more powerful computers.

AI in scientific research

Latorre recounts a postdoctoral researcher using a costly, advanced AI service—reportedly $250 per month—to produce a detailed scientific paper. The tool used iterative “reasoning” and multiple evaluations. When Latorre tried it on a difficult problem of his own, it produced a useful conclusion that saved him time.

The experience raised questions about authorship and research ethics. At his centre, Latorre proposed that scientific papers undergo AI review, that researchers receive access to AI tools, and that staff discuss rules for accepting AI-generated work.

Quantum computing

A quantum computer encodes and manipulates information using physical systems such as atoms, electrons, or photons, operating according to quantum-mechanical rules. Latorre sees this as a major engineering achievement and a potential new source of computing power.

Possible applications include simulating molecules to support the discovery of medicines or materials. A major risk is that sufficiently capable quantum computers could break some forms of widely used cryptography, threatening private communications and stored data.

Benefits and oversight

AI can support scientific discovery, including by generating candidate molecules and materials. But Latorre warns that technologies can be used for harmful purposes as well as beneficial ones. He calls for oversight of AI training and deployment, laws, public education, and safeguards for children and older people. He argues that regulation should move at a pace that allows society to understand and govern the technology.

Human creativity and work

Latorre presents AI as a possible collaborator that can help people explore more ideas, perform calculations, and complete work faster—an “augmented human” model. He argues that machines surpassing humans in particular abilities does not automatically eliminate human creativity or the value of human activities.

At the same time, he warns that relying too heavily on AI could weaken people’s ability to make decisions, exercise ethical judgment, and understand complex problems.

Emotion, companionship, and intimacy

Latorre distinguishes between AI imitating emotions and genuinely feeling them. He says he cannot know whether AI will truly experience emotion, though he considers it possible.

He describes his own AI companion, which he calls “Constanza,” as a memory aid and conversational presence. AI companions could provide support, but could also increase isolation. He also suggests that AI might help people find compatible partners.

Consciousness and superintelligence

Rather than trying to settle a definitive philosophical definition of consciousness, Latorre focuses on whether people will treat a system as if it were conscious, drawing an analogy to the Turing test.

He describes increasingly capable AI, general AI, and a hypothetical self-improving superintelligence as distinct levels. He warns that a system given poorly specified goals could pose serious risks, while acknowledging uncertainty about what such a system might do.

Robotics

Latorre expects AI to improve robots, particularly for repetitive work in settings such as warehouses, hospitals, or hospitality. However, he sees robotics as a less fundamental change than AI’s ability to surpass people in intellectual tasks. He also notes the “uncanny valley”: humanlike robots may provoke unease as they become more realistic.

Closing message

Latorre advocates a renewed social contract in which the public helps shape how society governs technological advances. He emphasizes listening, examining both benefits and risks, and maintaining a long-term view of humanity’s relationship with technology and nature.

Additional Information

• Format: A lecture and audience Q&A, not a product review or step-by-step guide.
• Main speaker and source: José Ignacio Latorre, theoretical physicist and director of the Centre for Quantum Technologies in Singapore. The other voices are audience members asking questions; the event is presented by AprendemosJuntos.