Meta has made a big AI language model available to the public.

Meta, which owns Facebook, is making a big language model available for artificial intelligence research, the social media giant announced on Tuesday.

Meta claims that its model is the first 175-billion-parameter language model available to the AI research community.

“Large language models” are natural language processing systems that have been trained on large amounts of text and can answer queries about reading comprehension or generate new content.

The publication of Meta’s “Open Pretrained Transformer (OPT-175B)” model, according to a blog post, will help researchers better understand how huge language models work.

Access to such models has been restricted, according to Meta, which has “hampered progress on efforts to improve their robustness and limit risk.”

Artificial intelligence technology, which is an important area of study and development for a number of large internet platforms, has the potential to perpetuate human societal biases on problems such as race and gender. Some scholars are concerned about the potential for huge language models to cause harm.

“We hoped to broaden the range of voices defining the ethical considerations of such technology,” Meta stated.

The IT giant said it was distributing the model under a noncommercial license to focus on research use cases in order to prevent misuse and “preserve integrity.”

Academic researchers, people linked with government, civil society, and academic groups, as well as corporate research facilities, will have access to the model, according to Meta. The models will be pre-trained, as well as the code to train them.