- The French parliament has validated the experimentation of algorithmic video surveillance on an experimental basis for the 2024 Olympics
- The government defends a necessary device to ensure security
- The opposition points to the risks of abuse of this technology
The controversy is not ready to end. At the end of last week, MEPs approved, on an experimental basis, the use of algorithmic video surveillance (VSA) to secure the Paris 2024 Olympic Games.
This device, which is part of a law dedicated to this event, was adopted by the few elected officials present in the hemicycle. 59 parliamentarians voted in favor (from the presidential majority, the National Rally, and the Republicans), and 14 against (New Popular Ecological and Social Union).
What is VSA?
There remains a question of size, what will this device be used for? This algorithmic video surveillance makes it possible to analyze images in real time. By relying on machine learning, AI advances over time. The idea is therefore to identify suspicious behavior, or crowd movements to react as quickly as possible.
Emmanuel Macron’s supporters notably claimed during the debates that this type of device could have foiled the Nice attack in 2016 by spotting a truck in the middle of the crowd. They also believe it could have helped the police maintain order during the last Champions League final at the Stade de France. The VSA would therefore be an additional asset for managing the millions of tourists who will visit Paris in the summer of 2024.
Risks of drift
Arguments that do not at all convince the opponents of the text. In a press release, the NGO, Amnesty International, pleaded as follows: “As France presents itself as a champion of human rights around the world, its decision to allow mass surveillance assisted by artificial intelligence during the Olympics will lead to a general offensive against the right to privacy, the right to protest and the rights to freedom of assembly and expression”.
Quoted by France 24one of its representatives, Katia Roux, specifies in particular: “These algorithms will be trained through a set of data decided and designed by humans. They will therefore be able to simply integrate the discriminatory biases of the people who designed and thought them..
Finally, the voted text risks being in opposition to future European legislation relating to AI. The latter effectively provides for a ban on the use of real-time remote biometric identification systems in publicly accessible spaces for law enforcement purposes. It incorporates certain exceptions, particularly in the face of “a specific threat” or in the case of an imminent risk of terrorist attack.
The legislative process is still following its course and the final adoption is scheduled for this Tuesday, March 28. An appeal to the Constitutional Council will then be filed, and the latter will decide whether or not to validate this controversial device.
