The beginning of new New AI model : OpenAI 
Image source: Internet
OpenAI, a global leader in AI, announced that its upcoming model is expected to deliver "the next level of capabilities" in their pursuit of creating artificial general intelligence (AGI)—a machine with human-like cognitive abilities. This new model will power a range of AI products, including chatbots, digital assistants like Apple’s Siri, search engines, and image generators.
“While we are proud to build and release models that are industry-leading on both capabilities and safety, we welcome a robust debate at this important moment,” the company said.
OpenAI is pushing the boundaries of AI technology faster than its competitors, while addressing concerns from critics who argue that AI poses risks such as spreading disinformation, displacing jobs, and even threatening humanity. Although experts debate the timeline for achieving artificial general intelligence (AGI), companies like OpenAI, Google, Meta, and Microsoft have been steadily enhancing AI capabilities for over a decade, showcasing significant advancements every two to three years.
Released in March 2023, OpenAI’s GPT-4 powers chatbots and various software applications, enabling them to answer questions, write emails, generate term papers, and analyze data. The updated version, unveiled this month but not yet widely available, adds the ability to generate images and respond to queries and commands with a highly conversational voice.
Shortly after the introduction of this updated version—dubbed GPT-4o—actress Scarlett Johansson claimed that the AI used a voice strikingly similar to hers. She revealed that she had refused OpenAI CEO Sam Altman’s offer to license her voice for the product and has since hired a lawyer, demanding that OpenAI cease using the voice.
Technologies such as GPT-4o develop their capabilities by processing extensive digital data, encompassing sounds, images, videos, Wikipedia entries, books, and news articles. In December, The New York Times filed a lawsuit against OpenAI and Microsoft, alleging copyright infringement related to AI systems.
The digital "training" of AI models can span months or even years. Following the training phase, AI companies usually dedicate several additional months to testing and refining the technology before it is released for public use.
This could indicate that OpenAI’s next model might not be available for another nine months to a year or more.
As OpenAI develops its new model, the newly formed Safety and Security committee will focus on refining policies and procedures to ensure the technology’s safe deployment. This committee includes CEO Sam Altman and OpenAI board members Bret Taylor, Adam D’Angelo, and Nicole Seligman. The company anticipates that these new policies could be implemented by late summer or fall.
In a notable shift, OpenAI announced this month that Ilya Sutskever, a co-founder and a key figure in its safety initiatives, is departing the company. This has raised concerns about whether OpenAI is adequately addressing the potential dangers of AI.
Dr. Sutskever had previously joined three other board members in November to oust Mr. Altman, citing doubts about his ability to responsibly manage the company’s mission to create artificial general intelligence (AGI) for the benefit of humanity. However, following a lobbying effort by Mr. Altman’s supporters, he was reinstated five days later and has since reasserted his leadership over the company. Company OpenAI confirmed the new model would be safe.
