OpenAI, an AI research and development company backed by Microsoft, has started releasing a powerful artificial intelligence model known as GPT-4. The model is multimodal, meaning it can generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way.
GPT-4 is the latest in a series of large language models developed by OpenAI. The company’s first model, GPT-1, was released in 2018, followed by GPT-2 in 2019 and GPT-3 in 2020.
GPT-4 is significantly larger than its predecessors, with 175 trillion parameters compared to GPT-3’s 175 billion. This means that it can generate significantly more complex and nuanced text.
The model has also been trained on a larger dataset of text and code, which gives it a better understanding of the world.
OpenAI, originally founded in 2015 as a non-profit organization and funded by Elon Musk, switched to a hybrid “capped profit” model in 2019, and is now worth over $30B. Microsoft remains the most significant investor in the company. While the intent was to originally democratize AI and counter other tech giants, it now has become closed-source and funded by one of the largest tech giants in the world.
ChatGPT-3 vs. ChatGPT-4
ChatGPT-4 and ChatGPT-3 (technically, GPT3.5) are both large language models (LLMs) that can generate text, translate languages, and write different kinds of creative content in response to a wide range of prompts and questions. However, there are some key differences between the two models.
ChatGPT-4 is trained on a dataset of text and code that is 175 billion times larger than ChatGPT-3’s dataset. This allows ChatGPT-4 to generate more coherent and informative text, and to translate languages more accurately.
ChatGPT-4 can also write different kinds of creative content, such as poems, code, scripts, musical pieces, email, letters, etc. It can also generate other creative text formats of text content, like poems, code, scripts, musical pieces, email, letters, etc.
In addition, ChatGPT-4 is multimodal, meaning it can understand and respond to images and videos. ChatGPT-3, on the other hand, was limited to text inputs and responses.
How to get developer access to GPT-4
OpenAI has said that GPT-4 will be available to developers through a waitlist. The company has also said that it will be used to power a number of new products, including a new version of ChatGPT, a chatbot that can hold conversations with humans.
OpenAI says it’s already partnered with a number of companies to integrate GPT-4 into their products, including Duolingo, Stripe, and Khan Academy. The new model is available to the general public via ChatGPT Plus, OpenAI’s $20 monthly ChatGPT subscription, and is powering Microsoft’s BingAI chatbot.
GPT-4 is a significant development in the field of artificial intelligence. It is the most powerful language model ever developed, and it has the potential to revolutionize the way we interact with computers.
However, there are also concerns about the potential misuse of GPT-4. For example, it could be used to create fake news or propaganda. It could also be used to generate harmful content, such as hate speech or pornography.
Already, ChatGPT-3 was able to generate malware code, social engineering email campaigns, and raise other cybersecurity concerns. GPT-4 does introduce additional safeguards, but as always, people will find a way to bypass AI safety guardrails.
Will GPT-4 give Microsoft a lead in the AI race?
Microsoft has already invested over $11 billion into OpenAI to power its AI ambitions, including building specialized supercomputers to process the multitude of daily BingAI chatbot requests.
Microsoft is so entrenched in OpenAI’s offerings that it is doubling down on their tech stack and just announced it will lay off its ethical AI team. Microsoft still maintains an active Office of Responsible AI, which continuously monitors and governs the company’s AI principles, however.
Google’s competitive ChatGPT equivalent, Bard, is still in limited testing with externally approved testers. Google CEO Sundar Pichai has asked every Googler to keep testing Bard, and is strategically inserting its Bard AI capabilities into solutions such as Google Workspace, to enhance productivity.
The AI race between big tech giants such as Microsoft, Google, and now even Meta are extremely interesting to watch develop in real-time. In many ways, this is analogous to the space race and nuclear arms race between the Soviets and the United States.
Article corrections: Corrected OpenAI description from non-profit to a “hybrid capped profit”; corrected GPT-4 to reflect 175 trillion parameters, not 175 billion parameters.
Disclaimer: The author of this article is a current employee of Google. This article does not represent the views or opinions of his employer and is not meant to be an official statement for Google, or Google Cloud.
