Viable Threat on News Reading: Generating Biased News Using Natural Language Models
This work highlights a threat to news aggregators and readers by showing how language models can be exploited to influence bias, though it is incremental in building on existing concerns about fake news.
The paper demonstrates that publicly available language models can reliably generate biased news content from original news, using controllable text generation to produce many high-quality articles, with evaluations showing 80 participants found them fluent and 24 participants could easily identify the bias.
Recent advancements in natural language generation has raised serious concerns. High-performance language models are widely used for language generation tasks because they are able to produce fluent and meaningful sentences. These models are already being used to create fake news. They can also be exploited to generate biased news, which can then be used to attack news aggregators to change their reader's behavior and influence their bias. In this paper, we use a threat model to demonstrate that the publicly available language models can reliably generate biased news content based on an input original news. We also show that a large number of high-quality biased news articles can be generated using controllable text generation. A subjective evaluation with 80 participants demonstrated that the generated biased news is generally fluent, and a bias evaluation with 24 participants demonstrated that the bias (left or right) is usually evident in the generated articles and can be easily identified.