Categoría: News

News

Scarlett Johansson vs. OpenAI: A Clash Over AI Voice Likeness

May 16, 2024 por jorge_carrasco 0 Comentarios #AI Ethics

Scarlett Johansson vs. OpenAI: A Clash Over AI Voice Likeness

In a world where artificial intelligence continues to blur the lines between reality and simulation, recent events involving Scarlett Johansson and OpenAI have spotlighted crucial ethical considerations. The acclaimed actress recently voiced her frustration and legal action against OpenAI for using a voice that bore an uncanny resemblance to hers in their latest ChatGPT 4.0 update, despite her previous refusal to participate.

Image generated by Artificial Intelligence (Stable Diffusion v1). Prompt: Joaquin Phoenix in the movie ‘Her’, sitting on a chair looking at the computer where Samantha is. In the style of a New Yorker magazine colourful illustration

The Incident Unfolds

Scarlett Johansson, known for her voice role as an AI in the 2013 film «Her,» revealed her dismay upon discovering that OpenAI’s new voice assistant, «Sky,» sounded strikingly similar to her voice. Johansson stated that she had been approached by OpenAI’s CEO, Sam Altman, nine months earlier to lend her voice to the ChatGPT 4.0 system, which she declined for personal reasons. Despite this, the voice of «Sky» released with the new ChatGPT 4.0 last week closely mimicked hers, enough to be mistaken by friends, family, and the public.

Presentation of new OpenAI Artificial Intelligence model, 3 people in front of an audience are using a phone where the new AI is. The room is made of wood. In the style of a New Yorker magazine colorful illustration

Johansson’s Statement

In her public statement, Johansson expressed feelings of shock, anger, and disbelief, noting that Altman had previously hinted at the potential comfort her voice could provide in bridging the gap between tech and creatives. The actor’s statement highlighted how this issue was not just about a voice but about the broader implications of consent, likeness, and the ethical use of AI technology.

«When I heard the released demo, I was shocked, angered and in disbelief that Mr. Altman would pursue a voice that sounded so eerily similar to mine that my closest friends and news outlets could not tell the difference,» Johansson wrote. This led her to seek legal counsel, resulting in OpenAI’s decision to pause the use of the «Sky» voice.

Image generated by Artificial Intelligence (Stable Diffusion v1). Prompt: Scarlett Johansson posing in the red carpet. In the style of a New Yorker magazine colourful illustration

OpenAI’s Response

OpenAI quickly responded by pulling the «Sky» voice from its ChatGPT 4.0 lineup. The company maintained that the voice was recorded by a professional actor and was not intended to mimic Johansson. «The voice of Sky is not Scarlett Johansson’s, and it was never intended to resemble hers,» Altman stated. He acknowledged that the voice actor was chosen before reaching out to Johansson, emphasizing that better communication might have avoided the controversy.

The Broader Implications

This incident underscores a growing concern within the entertainment and tech industries about the use of AI to replicate human likenesses. The rapid advancement of voice imitation technology allows for highly realistic reproductions, which can lead to significant ethical dilemmas and potential legal battles over likeness rights and consent.

Legal and Ethical Considerations

The case raises questions about the legal boundaries of voice and likeness imitation, particularly without explicit consent. Johansson’s call for transparency and appropriate legislation reflects a broader push for protecting individual rights in the face of advancing AI capabilities. This isn’t an isolated issue; voice imitation has been misused for scams and disinformation, highlighting the urgent need for regulatory frameworks.

Industry Reactions

The Screen Actors Guild – American Federation of Television and Radio Artists (SAG-AFTRA) voiced their support for Johansson, emphasizing the need for clarity and transparency. They commended OpenAI for pausing the use of «Sky» and expressed a desire to work with industry stakeholders to develop robust protections.

Conclusion

The Scarlett Johansson and OpenAI dispute is more than a celebrity grievance; it’s a pivotal moment in the ongoing dialogue about AI ethics and the protection of personal identity in the digital age. As AI continues to evolve, balancing innovation with respect for individual rights will be crucial. This incident serves as a reminder that with great technological power comes the responsibility to wield it ethically and transparently.

Stay tuned to our blog for more updates and insights on the intersections of artificial intelligence, technology, and ethical considerations.

News

Google Launches Gemini AI: What You Need to Know!

Ene 7, 2024 por jorge_carrasco 0 Comentarios #Deep Learning

Google Launches Gemini AI: What You Need to Know!

On December 6, Google unveiled its latest and most advanced AI model, Gemini, marking a significant leap forward in the realm of artificial intelligence. This groundbreaking model has already demonstrated its prowess by outperforming even the formidable GPT-4 in various benchmarks. In this blog post, we’ll delve into the key aspects of Google’s Gemini AI, exploring its capabilities, applications, and the expected impact it will have on the field of artificial intelligence.

Sundar Pichai, Demis Hassabis. (2023, 6 de diciembre). Introducing Gemini: our largest and most capable AI model. https://blog.google/technology/ai/google-gemini-ai/

So, What is Google Gemini?

Gemini stands as Google’s cutting-edge artificial intelligence model, designed to operate not only with text but also with images, videos, and audio. Setting itself apart as a multimodal model, Gemini showcases its ability to execute intricate tasks in fields like mathematics, physics, and beyond. Furthermore, it boasts the capability to comprehend and generate high-quality code in multiple programming languages.

Availability and Integrations

Presently, Gemini is accessible through integrations with Google Bard and the Google Pixel 8. Over time, it is slated to be seamlessly integrated into various other Google services. Notably, the most significant advancements in Gemini are anticipated in early 2024, coinciding with the launch of «Bard Advanced,» an enhanced version of the chatbot initially available to a select test audience.

Language Capabilities

Initially operating exclusively in English globally, Google assures that Gemini’s language capabilities will expand to encompass other languages in the future. This highlights Google’s commitment to ensuring the global accessibility and versatility of this powerful AI model.

Collaborative Development

Gemini is a collaborative creation of Google and Alphabet, Google’s parent company. The project also benefited significantly from the contributions of Google DeepMind, emphasizing the joint efforts of various entities within the Google ecosystem.

Different Sizes, Different Capabilities

Google Gemini is not a one-size-fits-all AI model; it comes in various sizes tailored to specific needs. Let’s explore the different versions:

1. Gemini Ultra

This is the largest and most potent model within the Gemini family, designed for highly complex tasks. While it is currently undergoing trust and safety checks and is available to a select audience, developers, partners, and safety experts, it is expected to be rolled out to a broader audience, including developers and enterprise customers, in the early months of the coming year.

2. Gemini Nano

Tailored for smartphones, specifically the Google Pixel 8, Gemini Nano is designed to perform on-device tasks efficiently without relying on external servers. Its applications include suggesting replies within chat applications or summarizing text.

3. Gemini Pro

Operating from Google’s data centers, Gemini Pro powers the latest iteration of Google’s AI chatbot, Bard. It excels in delivering rapid response times and understanding complex queries, making it a crucial component for enhancing user interactions.

Prapti Upadhayay. (2023, 6 de diciembre). Google Gemini vs OpenAI’s ChatGPT: Comparing the two most powerful generative AI tools. https://www.hindustantimes.com/technology/google-gemini-vs-openais-chatgpt-comparing-the-two-most-powerful-generative-ai-tools-101701883876150.html

Impact on the AI Industry

Google positions Gemini as a transformative force within the AI industry, distinguishing itself as the company’s most powerful AI model to date. Surpassing benchmarks set by OpenAI’s GPT-4, Gemini is poised to influence applications and devices significantly. Its initial deployment includes the Bard chatbot and Pixel 8 Pro, showcasing its versatility and potential impact on user experiences.

Google asserts that Gemini is one of the first models built as a multimodal large language model (LLM) from the ground up. This design choice aims to facilitate more natural and «human-like» interactions, further blurring the lines between man and machine.

The Road Ahead: Applications and Services

Google envisions Gemini extending its influence across various products and services. The model is expected to play a pivotal role in services like Search, Ads, Chrome, and Duet AI. Google has already initiated experiments with Gemini in Search, specifically in the Search Generative Experience (SGE). Early results indicate a 40% reduction in latency in English in the U.S., accompanied by improvements in quality.

As Gemini continues to evolve, it is anticipated to become a cornerstone in Google’s efforts to enhance user experiences across its diverse array of products and services. The integration of Gemini into various facets of Google’s ecosystem underscores the company’s commitment to pushing the boundaries of what AI can achieve.

Conclusion

Google’s Gemini AI represents a significant milestone in the evolution of artificial intelligence. Its multimodal capabilities, collaborative development, and diverse sizing options highlight its potential to redefine how AI interacts with and serves humanity. As Gemini gradually becomes available to a broader audience and integrates into more Google services, the impact on user experiences, efficiency, and the AI industry as a whole is likely to be substantial.

The launch of Gemini is not just a technological advancement; it’s a testament to Google’s dedication to pushing the boundaries of AI capabilities. As we step into this new era of AI with Gemini at the forefront, the possibilities for innovation and transformation seem boundless. Keep a close eye on Google’s Gemini for it is poised to shape the future of artificial intelligence in ways we are only beginning to comprehend.

FAQs

  1. What makes Google’s Gemini AI stand out in the world of artificial intelligence? Google’s Gemini AI distinguishes itself by being a multimodal model, excelling not only in text but also images, videos, and audio.
  2. How can developers access and utilize Google Gemini? Developers can currently access Gemini through integrations with Google Bard and Google Pixel 8, with broader access expected in early 2024.
  3. What languages does Google Gemini support, and are there plans for expansion? Initially operating in English, Google Gemini aims to expand its language capabilities globally, ensuring accessibility and versatility.
  4. Who were the key contributors to the development of Google Gemini AI? Google and Alphabet collaborated on Gemini’s creation, with significant contributions from Google DeepMind, showcasing a joint effort within the Google ecosystem.
  5. Can you explain the different versions of Google Gemini and their specific use cases? Google Gemini comes in various sizes, including Ultra for complex tasks, Nano for smartphones, and Pro for data center operations, catering to diverse needs.