How the new project works is a closely guarded secret, even within OpenAI. Some reports suggest a model with new human-like reasoning capabilities. This is what we know so far.
OpenAI, the developer of ChatGPT is working on a new approach to its artificial intelligence (AI) models in a project called “Strawberry,” according to a source familiar with the matter and internal documents reviewed by Reuters.
The project, details of which have not been previously announced, comes as the startup is backed by Microsoft is quick to demonstrate that the types of models it proposes are capable of offering advanced reasoning capabilities, like Artificial General Intelligence.
OpenAI teams are working on “Strawberry,” according to a copy of a recent internal OpenAI document seen by Reuters in May.
Reuters was unable to determine the exact date of the document, which details a plan for how OpenAI intends to use “Strawberry” to conduct research.
The source described the project to Reuters as a work in progress. The news agency could not determine how close “Strawberry” is to being publicly available.
What do we know about the Strawberry AI reasoning model?
How “Strawberry” works is a closely guarded secret, even within OpenAI, the source told Reuters.
The document details a project that uses “Strawberry” models with the goal of enabling AI to not only generate answers to queries, but also to plan an internet navigation well in advance in an autonomous and reliable manner in order to do what the company called “deep investigation,” according to the source.
It’s something AI models have so far overlooked, according to interviews with more than a dozen AI researchers.
Asked about “Strawberry” and the details reported in this article, an OpenAI spokesperson said in a statement: “We want our AI models to see and understand the world more like we do. “Seeking new AI capabilities is a common practice in the industry, with the shared belief that these systems will improve their reasoning over time.”
The spokesperson did not directly respond to questions about “Strawberry.” The “Strawberry” project was previously known as “Q*” and, according to Reuters last year, was already considered a major breakthrough within the company.
Two sources described seeing earlier this year what OpenAI staff told them were demonstrations of “Q*,” which can answer difficult questions in subjects like science and math that are outside the scope of currently commercially available models.
On Tuesday, at an internal meeting, OpenAI demonstrated a research project that it said had new human-like reasoning capabilities. Bloomberg .
An OpenAI spokesperson confirmed the meeting but declined to provide details on its content. Reuters could not determine whether the project presented was “Strawberry.”
OpenAI expects the innovation to significantly improve the reasoning capabilities of its AI models, the source said, adding that “Strawberry” involves a specialized way of processing an AI model after it has been pre-trained on very large datasets.
Source: Latercera
I am David Jack and I have been working in the news industry for over 10 years. As an experienced journalist, I specialize in covering sports news with a focus on golf. My articles have been published by some of the most respected publications in the world including The New York Times and Sports Illustrated.
