If the trend continues, artificial intelligence could “benefit only a small, homogeneous group” of people within our societies.
The warning comes from Miriam Vogel, who chairs the US National Advisory Committee on Artificial Intelligence – whose mandate is to advise the US president on issues related to AI.
She sounded the alarm in early July, in an interview with the news site Axios.1where she argues that “the only thing we have to fear from AI is the fear of AI itself.”
She says she found that “despite the enthusiasm of investors, tech companies and some early adopters, the vast majority of people are still afraid to use AI.”
Among other things, she advocates the establishment of artificial intelligence initiation programs “to encourage more people to take an interest in the technology” and, at the same time, help them develop their critical thinking towards it.
I am talking about Miriam Vogel’s public statement here because this warning is not limited to the United States.
There is also a digital literacy problem in Quebec that needs to be addressed. This is common knowledge.
I spoke about it with several experts for this column, starting with Yoshua Bengio, full professor at the University of Montreal, but also founder and scientific director of Mila, the Quebec Institute of Artificial Intelligence.
PHOTO JOSIE DESMARAIS, LA PRESSE ARCHIVES
Yoshua Bengio, full professor at the University of Montreal and founder and scientific director of Mila
“It’s certainly better than it was, because we’ve been talking about AI in the last year like we’ve never talked about it before,” he says of AI literacy.
However, it remains insufficient.
“At the same time, there are so many misconceptions, including fears that are not based on scientific thinking,” he adds. “There will be political decisions to be made. And it’s a bit like with the climate, people need to understand the scientific basis of the issues and the risks so that they can push governments in the right direction, according to what science and our value choices lead us to make as decisions.”
Last fall, at the first global summit on AI risks in the United Kingdom, Yoshua Bengio was tasked with chairing an international scientific report on the issue. The preliminary report was released a few weeks ago.
I asked this expert why it is so important for everyone to become familiar with artificial intelligence.
“Because there is a good chance that AI will change the world in ways that are both positive and dangerous,” he replied. “And to be able to deal with it, to take advantage of it where it is beneficial and to avoid disasters, we need as many people as possible to understand the basis, the risks, how it works. What could be useful for me, in my work, in my life, but also what could be used against me.”
The idea that artificial intelligence could benefit only a small, homogeneous group of people has provoked a strong reaction from Lyse Langlois, general director of the International Observatory on the Societal Impacts of AI and Digital Technology (OBVIA).
“This has been a concern for us from the beginning. You should know that even before the creation of the Observatory, in 2019, there was research that showed the impacts in terms of bias, inequalities, manipulation… And also, when I think of disinformation among other things, the fact that there was a lack of literacy and knowledge,” she emphasizes.
Many researchers have put forward the idea that critical thinking and digital literacy should be our priorities.
Lyse Langlois, General Director of OBVIA
Important clarification: this does not mean that nothing is currently being done in Quebec on this subject.
Lyse Langlois explains that OBVIA “is taking a lot of initiatives.” She also points out that the Quebec Innovation Council launched a dialogue on this subject last year when it published its report on the supervision of artificial intelligence.
Another expert who thinks it’s important to become more familiar with artificial intelligence is digital literacy professor Julie Corrigan of Concordia University.
“If we think about any technology… A snowplow, for example. There used to be teams of men who would go around and clear the roads. So let’s say I don’t care about the snowplow and I just decide to ignore this technology. I would be out of a job,” she says.
What we need to understand is when to use AI, how to use it, and why AI can be an advantage or a disadvantage.
Julie Corrigan, professor specializing in digital literacy at Concordia University
I conclude by pointing out that while Yoshua Bengio is generally on the same wavelength as Miriam Vogel, he insisted in an interview that she is wrong on the question of fear.
She: Fear of AI is a problem.
Him: It would not be rational to fight against fears about AI, because they are well-founded.
“Yes, we should be afraid of AI, but not the AI that is in commercial products today,” he says.
“We should be afraid of AI research, which is being done in labs like OpenAI, which aims for systems that are as intelligent or more intelligent than us in several years. That’s what’s dangerous. Because scientists don’t know how to develop them in a way that they can’t be used as a dangerous weapon or that they don’t turn against humans.”
In short, it is legitimate to be afraid of artificial intelligence. The problem is to start with fear and, for this reason, to stay away from AI. In this area, fear can be both a good… and a bad advisor!
1. Read Miriam Vogel’s interview with Axios
What do you think? Join the dialogue