Technology: The organization, which has been in existence for five years, is launching its first commercial product, allowing companies to access its multi-purpose AI models.
OpenAI is launching its first commercial product, the company announced Thursday,which gives companies access to its most advanced general-purpose AI models via an API. The API, launched in private beta, is currently used by customers for a range of applications, including semantic research, sentiment analysis and content moderation.
While most AI models are designed for specific use cases, OpenAI’s API provides a general-purpose text-in, text-out interface that could be applied to a wide range of English-language tasks.
The API performs models with weights from the GPT-3family, the family of massive neural networks of OpenAI. The recently released GPT-3 uses 175 billion parameters,allowing it to achieve “meta-learning” – meaning that the GPT neural network is not re-trained to perform a task such as completing sentences.
Reddit explores content moderation with API
If you give the new API a typing line, it will attempt to return a completed text that matches the template given to it. Users can fine-tune its performance on specific tasks by training it on a set of examples, large or small, or through human comments provided by users or labelers.
Early customers include the Quizlet online learning platform, which uses the API to automatically generate examples of vocabulary word use in a sentence. Reddit explores content moderation with the API, while the legal research platform Casetext aims to improve its semantic search capabilities with it. The MessageBird cloud communication platform uses the API to develop automated spelling and grammar tools, as well as predictive text capabilities.
OpenAI was founded in 2015 by Sam Altman, former president of the Y Combinator, and Elon Musk, CEO of Tesla. The research and deployment company focuses on “artificial general intelligence,” which it defines as “highly autonomous systems that outperform humans in the most cost-effective jobs.”
Preventing harmful use cases
The company is launching its private beta API in part because of the risks associated with launching a versatile AI tool.
“We will end access to the API for obviously harmful use cases, such as harassment, spam, radicalization or astroturfing,” reads the OpenAI blog. “But we also know that we cannot anticipate all the possible consequences of this technology.”
In addition to limiting its availability, OpenAI stated that it was developing tools to help users better control the content returned by the API, and that it was studying security-related aspects of language technology, such as analysis, mitigation and intervention in the event of adverse bias.
Source : ZDNet.com