OpenAI conducted its first-ever developer conference in San Francisco. As expected, there were a lot of announcements about the company’s new version of its large language model (LLM) named GPT-4 Turbo. Sam Altman, the CEO of OpenAI called Microsoft CEO Satya Nadella to the stage to talk about the partnership between both the companies.
Altman initiated a conversation about the ongoing partnership between Microsoft and OpenAI. In response to Altman’s question about Microsoft’s perspective on the partnership, Nadella responded with a hearty laugh, “We Love you guys. You guys have built something magical.”
Nadella went on to elaborate on two key aspects of the partnership. The first aspect he highlighted was the unique and novel workloads that the partnership has been handling. He described the training jobs as synchronous, large, and data-parallel. “The first thing we’ve been doing is building the partnership with the system. The shape of Azure has drastically changed and is changing in support of these models that you’re building,” Nadella said.
He emphasised that their primary goal is to build the best systems so that OpenAI can build the best models and make them available to developers.
The second aspect Nadella touched upon was Microsoft’s identity as a developer. He shared his conviction about the potential of foundation models, which was significantly influenced by GitHub Co-Pilot on GPT. “We want to build our co-pilot as developers on top of OpenAI API,” Nadella stated.
He further announced that they aim to make GitHub Co-Pilot, enterprise edition available to all the attendees. He assured the audience that they can count on Microsoft to build the best infrastructure in Azure with OpenAI API support and bring it to all developers. He also mentioned plans for the Azure marketplace to facilitate developers in rapidly bringing their products to market.
OpenAI Announcements
OpenAI’s Sam Altman made a lot of new announcements at the DevDay conference. Here’s a look at the six top announcements made by the company:
1. GPT-4 Turbo: This is the latest model in OpenAI’s GPT series. It has a larger context window of 128,000 tokens and has improved world knowledge compared to its predecessor. It can process images as input and generate human-like speech. There are two versions: one that works with text only, and another that works with both text and images.
2. Costing of new Model: GPT-4 Turbo is designed to be more cost-effective than its predecessor. The cost for input tokens will be reduced by three times, and the cost for output tokens will be halved.
3. ChatGPT Update: The knowledge base of the ChatGPT bot is being updated. Previously, it was limited to information up to September 2021. With the update, it will have information up to April 2023.
4. Copyright Shield: OpenAI has introduced a “copyright shield” to protect its customers. This means that if a user generates text using OpenAI’s GPT Enterprise API and faces a copyright infringement lawsuit, OpenAI will provide protection.
5. Custom ChatGPT: OpenAI is developing a feature that will allow developers to create custom chatbots, or GPTs, tailored to specific use cases.
6. GPT Store: OpenAI announced a new platform called the GPT Store. However, the specifics of what this platform will offer were not disclosed at the conference.
Also read: OpenAI has released a new tool that will let you build your own ChatGPT; see how it works
Also read: Sam Altman’s OpenAI unveils new and improved GPT-4 Turbo: All you need to know
VIDEO
{Categories} _Category: Takes,*ALL*{/Categories}
{URL}https://www.businesstoday.in/technology/news/story/we-love-you-guys-microsoft-ceo-satya-nadella-shares-stage-with-sam-altman-at-first-openai-developer-conference-404823-2023-11-07{/URL}
{Author}unknown{/Author}
{Image}https://akm-img-a-in.tosshub.com/businesstoday/images/story/202311/untitled_design_-_2023-11-07t110557-sixteen_nine.jpg{/Image}
{Keywords}{/Keywords}
{Source}Platforms{/Source}
{Thumb}{/Thumb}