Chatbots are in fashion, and they were born yesterday, so to speak. Today the whole world is seeing how to take advantage of this technology. The student believes that it will facilitate the preparation of school or college tasks. The reporter thinks that in a few minutes he will already have the base of his article. The groom is in the task of having the chatbots compose a poem to give it to the bride as if it were his, etc.
Those who receive what is written will from now on have the arduous task of deciphering if what they are reading has a personal or Chatbotian authorship, because from the outset, we will all have doubts as to whether what we are reading is the result of Intelligence Natural, or artificial intelligence.
The problem with new chatbots is not just that they are often or seem stupid and naive; it is that they are not “stupid” or “naive” enough to grasp the nuances of language, the ironies, and the revealing contradictions that constitute human culture and communication. They look like us, but they are not us. So by trusting them, we run the risk of succumbing to the same stupidity.
It seems that the support technology, which is already on version 4, GPT-4, is much more powerful than the technology of previous versions and much more capable, and therefore more truthful in appearance, which does not mean that it is.
Now, it will not only make us doubt what we hear or what we read, but also what we see that is not live. Seeing is believing, it’s over. Now we also have to sift what we see and hear. Deepfakes are videos manipulated to make those who see them believe that they are seeing a certain person making statements that never took place, but that someone else constructed. For the manufacture of these files, artificial intelligence tools or programs are used that allow the exchange and manipulation of faces in live images, together with the transformation of the voice, to such an extent that the AI is capable of transforming a photo into a talking image that addresses us from a screen.
Worst of all, it is not so easy to decipher what is a deepfake of a real face in an image that is speaking to us.
To differentiate a deepfake from a real face, according to the expert Chema Alonso, one must know how to discover certain subtleties of the image that become very noticeable in blinking: “We study how much the person blinks and what distance there is between blinks.” We capture all the gestures, the iris and the reflection of light are also studied. It is important to observe the heartbeat that is reflected in the face”, it is not something that is within everyone’s reach. Such is the capacity of these new technologies that are at the beginning of their development, for which a world of evolved.
With these technological advances placed at the service of the human being, chatbots may be smarter and better enough to replace us in what we do, but at that very moment humans will be more stupid and worse, I don’t know if due to the fact delegating our natural intelligence and its oral and written production to the equivalent artificial intelligence, we will suffer a regression of our capabilities.
Nature teaches us that what is not used atrophies because it is no longer used, and it is left over. It may be that the machines continue their progress and are increasingly capable, at the same time that living beings regress towards more primitive stages. If you don’t walk, your bones weaken and your muscles lose mass, and with them your strength goes away. Well, with Natural Intelligence and the use we make of it, I think the same thing would happen if we loosen up in its use. We better keep it in mind if we don’t want to stop being Homo Sapiens. We will continue to be Homo, but the adjective Sapiens will have to be replaced by another word that is better suited to what we would be if we stopped using our brain and its neurons.
El artículo se puede leer en español en este enlace.
Comments