OpenAI Releases GPT-4 – Unlocking The Potential Of Artificial Intelligence

On Tuesday, OpenAI, the San Francisco-based technology business renowned for introducing ChatGPT, announced the launching of a new version of its AI program.

OpenAI has noted that GPT-4, the software they have developed, can accomplish difficult tasks with a higher degree of accuracy due to its great general knowledge and problem-solving skills. This was declared in an announcement on their website.

In a video showcasing OpenAI’s technology, Greg Brockman, the company’s president, demonstrated its use in quickly responding to tax-related questions like determining the standard deduction and total tax liability for a married couple.

Greg Brockman says:

“This model is so good at mental math.”

“It has these broad capabilities that are so flexible.”

In a different video put up by the corporation on the internet, they stated that GPT-4 possessed various features not present in its preceding version. One of these was the capability to “make inferences” on images provided by people.

OpenAI states on its website that GPT-4 is a sizeable multimodal model, able to take in both image and text inputs and return text outputs. Although it may not perform as well as humans in many everyday situations, the model has been shown to reach human levels of accuracy when tested against professional and academic benchmarks. Cy when tested against professional and academic benchmarks.

Andrej Karpathy, an OpenAI employee, tweeted that the AI had been able to “see” with the new feature. Unfortunately, this technology is not available for free yet; however, people can try out GPT-4 on OpenAI’s subscription service ChatGPT Plus which is priced at $20 per month.

OpenAI’s ChatGPT chatbot has impacted the tech world and caused many people outside of the industry to become aware of AI software’s potential. This is partly due to OpenAI’s collaboration with Microsoft and Bing search engines.

The accelerated progression of OpenAI’s releases has sparked worry as the technology is yet to be properly assessed, thus leading to a sudden shift in many sectors, such as education or the arts. This fast-track public growth of ChatGPT and other generative AI systems has caused some ethics experts and industry professionals to call for restrictions on using this type of technology.

On Monday, Sam Altman, CEO of OpenAI, tweeted that there is a need for increased regulation of artificial intelligence. To demonstrate the capabilities of GPT-4, the company provided examples on its website, which included booking meetings with three busy people without any conflicts, achieving high scores on tests such as the uniform bar exam, and being able to mirror a user’s writing style.

Google released its software, Bard, in February due to concern that Artificial Intelligence (AI) technology could impact the market share of its search engine and cloud-computing service. However, the company also acknowledged potential drawbacks such as social biases and “hallucinations” where it believes it knows more than it does.

In 2015, OpenAI was founded with support from notable names such as Elon Musk, Peter Thiel, Reid Hoffman, and other tech moguls. The company’s name symbolized its status as an open-source project that would provide software freely over the web. In 2019, it shifted to a “limited” profit-oriented business structure.

The release of GPT-4 has been done with some degree of secrecy. An accompanying 98-page paper stated that many details would be kept under wraps, including the data used to train the model, which will not be revealed publicly.

This report avoids presenting further information regarding the architecture, hardware, training compute, dataset development, training technique, or related topics due to the competitive environment and the safety considerations presented by large-scale models like GPT-4.

They declared that they would furnish more technical information to extra outside parties who can help them assess the competence and safety aspects against the scientific benefit of increased openness.

Rumors of the fourth version of OpenAI’s core system, GPT-4, have been circulating for months as enthusiasm grows over the chatbot that is based on it. In January, Altman attempted to downplay people’s anticipation of what GPT-4 could do by saying on StrictlyVC that “they are pleading to be let down and they will be.”

On Tuesday, he solicited feedback.

Altman says:

“We have had the initial training of GPT-4 done for quite awhile, but it’s taken us a long time and a lot of work to feel ready to release it.”

“We hope you enjoy it and we really appreciate feedback on its shortcomings.”

Sarah Myers West, the managing director of the AI Now Institute, a nonprofit group that studies the effects of AI on society, said releasing such systems to the public without oversight.

Altman went on to say:

“Is essentially experimenting in the wild.”

“We have clear evidence that generative AI systems routinely produce error-prone, derogatory and discriminatory results,” she said in a text message. “We can’t just rely on company claims that they’ll find technical fixes for these complex problems.”

Despite these challenges, the release of the GPT-4 represents a major milestone in the development of artificial intelligence. It is set to unlock new and exciting possibilities for businesses, researchers, and individuals worldwide. As we continue to explore the potential of AI and push the boundaries of what is possible, the GPT-4 will undoubtedly play a critical role in shaping the future of technology and transforming the way we live and work.

Source: NBC News

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top