Site navigation

OpenAI to Launch Text Tool Previously Deemed Too Dangerous

David Paul

,

OpenAI

The Elon Musk-backed machine learning company is releasing the software as a private beta rather than open-source after previously claiming it could be misused.

Software company OpenAI, co-founded by SpaceX owner Elon Musk, is launching its text generation software in a private form after previously saying it was too dangerous to release.

The rentable text generation tool can take a short sentence and generate a full story using the tone of the text. Once it is given any text prompt, it will return a text completion, attempting to match the pattern given to it.

It is also possible to ‘programme’ the software by showing it examples of what you would like it to do, with its success depending on the complexity of the task.

So far it has taken the opening line of George Orwell’s Nineteen Eighty-Four and extended it into a more fleshed out story, as well as re-writing a Guardian article.

The company now says it wants to utilise the technology in coding and data input, potentially reducing the man-hours needed for mundane data entry.

The technology was originally intended to be released as open-source, but OpenAI deemed the potential for it to be misused as dangerous.

Recommended

OpenAI said in a blog: “The field’s pace of progress means that there are frequently surprising new applications of AI, both positive and negative.

“We will terminate API access for obviously harmful use-cases, such as harassment, spam, radicalisation, or astroturfing.

“But we also know we can’t anticipate all of the possible consequences of this technology, so we are launching today in a private beta rather than general availability, building tools to help users better control the content of our API returns, and researching safety-relevant aspects of language technology (such as analysing, mitigating, and intervening on harmful bias).

“In releasing the API, we are working closely with our partners to see what challenges arise when AI systems are used in the real world.

“This will help guide our efforts to understand how deploying future AI systems will go, and what we need to do to make sure they are safe and beneficial for everyone.”

David Paul

Staff Writer, DIGIT

Latest News

Infrastructure Trending Articles
Cybersecurity Editor's Picks Government
Events Featured Social Media
%d bloggers like this: