Some people are calling for the application of the right to free speech for artificial intelligence. A way of supporting freedom of thought, but which could have the opposite effect depending on the case.
Whether we like it or not, theartificial intelligence is here to stay. Increasingly present in our daily lives, it fascinates as much as it worries the majority of people. This rise in power has not escaped the attention of institutions and governments which have been seeking to regulate it for some time. Generally speaking, give a legal framework for AI would allowavoid abuse, even if some see it as a brake on innovation. But the establishment of such a framework raises a fundamental question: should artificial intelligence be able to express itself freely ?
For some experts, the answer is yes. Their reasoning is as follows: the right to freedom of expression is inalienable and contributes to another fundamental human right, freedom of thought. Artificial intelligence, by giving us answers and providing us with information, support our thinking. And to do this, AI should not be limited in its speech.
Experts say AI should benefit from freedom of expression
In the United States, the Supreme Court clearly states that freedom of expression “does not depend on the identity of its source”. Whether it is a person, a company or an AI, the important thing is that it participates in present a variety of ideas. We can clearly see the risk here: the disinformation. AI could become a source of rumors and false information if it were free to “say” whatever it wants. At the same time, preventing it from doing so would be seen as an attack on freedom of thought: if AI can only give one type of information, it actually guides the people who use it. It is then up to them to sort things out, in one case or the other.
Read also – The CNIL unveils its plan to supervise Artificial Intelligence, here are the main points
The question of whether or not to grant freedom of expression to artificial intelligence is complex. In its regulations, Europe requires that content generated by AI be clearly identified as such. It must not be illegal, and companies must mention a summary of sources used for training of artificial intelligence. This could help qualify his remarks. However, nothing says that future legal frameworks will integrate the notion of free expression and if so, in what form.
Source : The Conversation
