LONDON (Reuters) - The British public broadcaster and the world's largest audiovisual and online broadcasting network, the BBC, is drawing up plans to create and train its own artificial intelligence model using its text files.
See also: Nvidia and Hippocratic AI collaborate to bring generative AI nurses
A BBC executive told a parliamentary committee earlier this month that the organisation is looking at genetic artificial intelligence as a tool to support its production process and to assess a potential unilateral or co-production scheme for training a model.
Regarding a Financial Times on Thursday that the BBC intends to create its own artificial intelligence models, a spokesperson confirmed that the BBC is looking at developing a Large Language Learning Model (LLM) using text-based content.
See also: Neuralink: Patient with brain chip plays chess with his thoughts
"The power derived from such a model would only be available on the BBC network," the spokesman said.
The Financial Times also reported, citing sources, that the television network is in discussions with technology companies about selling access to its archives, which could then be used to train artificial intelligence models.
See also: Asteroid Bennu: Is there a chance it will eventually hit Earth?
Responding to this, the BBC said: “The BBC has no agreement with any organisation to use its records to train the large language learning models that power genetic AI tools for commercial use.”
"We are looking at a wide range of issues such as the potential bias within such models and how the BBC, either collaboratively or unilaterally, can address these."
Source: finance.yahoo.com
