There is an urgent need to regulate AI, but is it possible?

There is an urgent need to regulate AI, but is it possible?

News media and tech companies may be constantly reiterating that artificial intelligence is the future, but it is important to remember that it also poses serious dangers. These are not Terminator scenarios where the machines take over the world, but rather the good old scenario where the human could do anything with the machine. Weapons management will be one of the most problematic issues.

AI can theoretically change the world as we know it. One of the most positive consequences would be for AI to cure cancer, for example, but as always you have to accept the other side of the coin: it changes the way we wage war and its use sometimes raises social, ethical and even political issues.

AI robot 02
Can artificial intelligence help or save humanity? / © Jirsak/Shutterstock

Borrowing a JFK quote, we could say that the current situation can be summed up as "don't ask yourself what AI can do for you but what you can do for AI", because, as François Hollande would say, "change is now". Limits must be set now if we don't want the situation to go out of control. This technology "could be dangerous for capitalism and for democracy itself", says Forbes. That is the least we can say, because AI is supported by the largest economic actors, strongly interested in its particularly lucrative aspect. It has already successfully interfered in many public services, including medicine and national security.

The most convincing example of the need for regulation is undoubtedly that of armaments. A former US Secretary of State, Henry Kissinger, concretely explained that "it may be a lot harder to control the development of AI weapons than nuclear ones". This sentence says a lot about the approach toward this technology. We must not forget that we must not let ourselves be lulled by new tech and its big actors who want us to believe that everyone is nice. The use of AI is not only pacifist and the need for regulation is urgent.

Forbes says that "Artificial Intelligence regulation may be impossible to achieve without better AI, ironically." In other words, machines should control humans. A very sad (or strange, depending on the point of view) conclusion.

What is your opinion on the matter? Let us know in the comments.

Source: Forbes

Liked this article? Share now!
Join the discussion


Write new comment:
All changes will be saved. No drafts are saved when editing
Write new comment:
All changes will be saved. No drafts are saved when editing

  • storm Mar 5, 2019 Link to comment

    No, that genie is out of that bottle. It will require consumers to abandon surveillance capitalism first and pursue open source hard and soft solutions on their own. Once there is public review of what we're using and we can see the abuses, then we'll make different choices in our spending and use.

    Case in point, Facebook. Google to a lesser degree.

    Deactivated Account

  • Albin Foro Mar 4, 2019 Link to comment

    I'd like to see a third-party standard measure of the IQ of AI, a Stanford Binet test of sorts. The ballyhooed Amazon AI, which has thousands of dollars over 20 years of household purchases from tech to groceries, coughs up nothing but nonsense: if I order two packages of four boxer shorts, it immediately and brilliantly "recommends for me" buying even more packages of same at a higher price. Imagine a high school drop out sales clerk in a department store doing that? My nightmare is not what AI can do, but that a lot of grinning idiots in high places will use government agencies to obtain and then use government police powers to enforce AI "recommendations" by laughing algorithms.