Home » Sciences et technologies » forum.eu | Openais mimetic masks

forum.eu | Openais mimetic masks

OpenAI et Scarlett Johansson : une polémique sur l’IA vocale

PARIS – 21 Mai 2024 –

La controverse enfle autour d’openai suite à l’utilisation présumée de la voix de Scarlett Johansson dans son nouveau chatbot. L’actrice,par le biais d’une déclaration publique,dénonce l’usage non autorisé de sa voix artificielle et annonce des mesures légales. Cette affaire soulève des questions éthiques cruciales sur l’utilisation par OpenAI de l’image des personnalités dans le domaine de l’IA, notamment concernant la ressemblance vocale.

2024-05-21 11:55:00

Last week, Openai presented the latest Chatgpt update: Chatbot can now “talk” and “conversation” in different voices, completely with “human” fill words (“uh”, “uh” etc.), breaks, simulated breathing, laughing and giggling. This voice of the new Chatbot version sounded just like that of Scarlett Johansson, who already synchronized a AI-for Spike Jonze’s film “Her”, a melancholic character study on the consequences of emotional bonds on AI systems. During the presentation still tweeted (XTE?) Openai-CEO Sam Altman a single word: “Her”.

Today Scarlett Johansson published a statement in which she announced legal steps and opened Openaai the voice option “Sky” from Chatgpt. Everything is just a misunderstanding and the similarity to the voice of Scarlett Johansson. Clearo. As Johansson’s statement shows, Openai and the actress had been in negotiations for months, and two days before the presentation, Altman tried to move the actress to lend her voice to his “her” version. Johansson refused to consent, and Openaai still used a voice color that was the maximum similar to that of Johansson.

I would like to use this episode of the unscrupulous handling of AI companies with personality markers of people to point out a text about Bernard Stiegler‘s philosophy. Stiegler bordered the concept of “technology” from the term “technology” to show how our fascination with the surface of the technologies – the shiny gadgets, the colorful interfaces, the sleek forms, and today: the seductive voices of well -known actresses – our actual relationship to the tool and its actual effect on us as a person in our socially social structure.

Stiegler developed the concept of “pharmacology”, according to the technologies of new media, act similarly to us such as pharmaceuticals, and can have good and harmful effects, similar to medicine. Stiegler’s concept of technology indicates exactly these “pharmaceutical” effects of technology: the “side effects” that slumber under the surface of the interfaces and pure application results.

Sam Altman gave the latest version Chatgpt’s voice of a well -known actress with the explicit goal of actual realization of the fictional chatbot from Spike Johnson’s film. The question is now to argue with a step, which “pharmacological” effects actually hides behind this Scarlett-Johansson interface. Two things come in spontaneously.

On the one hand, a study recently confirmed the psychoactive effect of AI chatbots: An experiment with over 2000 supporters of conspiracy theories showed that entertainment with a chat bot was able to reduce the beliefs of these people by around 20%. That sounds superficially great, but at first it simply means “only” that a chat bot is able to influence people so far that they change fundamental attitudes. AI systems are apparently in the word sturn convincing -and the use of the familiar voice of Scarlett Johansson forms a direct connection between the acoustic, now also Hollywood glamorous interface interface of the chattt technology and the translucent seduction of the language-simulating technology.

(A year ago I signed the open moratorium letter for this reason: I am a little afraid of alleged Ai-Doomsday scenarios. But the seduction of mimetic and anthropomorphic AI systems that suggest that the simulated language is “personality”, and the undiabaric psychological consequences if these “personalities” lead to a synthetic theory, in the AI ​​systems one real and effective Take space as “social actors” in our individual psyche, worries me more than any robotic office clip-AI handicraft phillip. Another, similarly stored example would be the psychoactive effect of romantic chatbots that fulfill all our wishes on the command, and can therefore produce unforeseeable and unfulfilled demands in relationships with humans.)

By using the (non-consensual) voices of well-known actresses, and she emphasizes “humanly” (ie filler words and ohs and Ahs and giggles), Openai tries to achieve “under the Skin” of her users (to cite the title of another film with Scarlett Johansson). With this AI-mimetic seduction tactic, Openai on the one hand reveals the psychoactive effect of her product, on the other hand, of course, the fact that we pay expensive for this synthetic seduction through a chat bot in the form of gigantic CO2 balance sheets and water consumption of datacenters, which increases by 30% for Microsoft in 2022 alone.

These are just a few of the possible and actual effects of chatt Technologythe Openais Techno-mimetic masks are already disguising today. More will follow.


#forum.eu #Openais #mimetic #masks
1748958619

You may also like

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.