The introduction of GPT-4, the improved version of Chat GPT, in March of this year, was still fresh news when industry experts had already hinted at the development of GPT-5. Concerns and dangers surrounding this type of AI had already raised alarms across the globe since the release of Chat GPT (version 3.5 of GPT). In late March, thousands of AI experts, including Elon Musk and Apple co-founder Steve Wozniak, signed an open letter calling for a six-month pause in the development of these AI systems. The goal? To develop and implement a set of shared safety protocols, to reflect on necessary regulations, and to establish safeguards before allowing AI labs to continue in an uncontrollable race.

While the CEO of OpenAI, Sam Altman, denied these rumors and stated during a conference at MIT, We are not there and wont be there for some time, the GPT-5 trademark was registered on July 18th. Siqi Chen, CEO of several tech companies, also declared on social media, Ive been told that GPT-5 should finish its training in December, and OpenAI expects it to achieve AGI [Artificial General Intelligence].

GPT-4, the latest version of Chat GPT, has reportedly improved its factual accuracy by 40% across all evaluated categories such as math, history, science, and writing, according to OpenAI. It is now close to reaching 80% accuracy in its responses. Experts believe that GPT-5 will surpass the 90% accuracy mark.

The major advancement in the latest versions of GPT is the Multisensory AI Model. While Chat GPT only deals with text, GPT-4 can process both text and images. Experts expect that GPT-5 will have the ability to process multisensory data, including audio, video, temperature, and other forms of data.

The question remains: will GPT-5 achieve Artificial General Intelligence? OpenAIs CEO, Sam Altman, has previously described how AGI could benefit humanity but has also warned about the dangers it poses. I think if this technology goes wrong, it can go really wrong. And we want to be out there, very loudly and clearly, saying this is risky. We want to work with the government to prevent that from happening, he declared during a hearing at the United States Senate.

While only Siqi Chens statements suggest that GPT-5 could reach AGI, the trademark registration serves as a warning of its inevitable release in the coming months. As the competition intensifies among tech giants like Google, Apple, Facebook, and Microsoft in the chatbot technology race, the prevailing question remains: will it (soon) achieve Artificial General Intelligence? Or will regulations and safety protocols be in place beforehand?

Original post:

Development of GPT-5: The Next Step in AI Technology - Fagen wasanni