Intel’s new AI helps you get the right amount of hate speech in chat games

Intel Microprocessor was founded in 1968. It has gone through a technological and innovative trail for decades after becoming one of the world’s leading manufacturers of computer chips.

But it doesn’t matter. Because we live in a world where Kodak is a failed cryptocurrency company that now deals in drugs and everyone still thinks Elon Musk invented the tunnel.

Which means that in this darkest time sequence, we’re stuck with a version of Intel that uses AI to run the “white nationalism” slider and the “N-word” switch for video game chat.

Watch “Sleep” in all its stupid glory:

What you see is the user interface for Bleep. An artificial intelligence-based software solution featuring a range of sliders and switches that allow you to determine how much hate speech in a particular category you want to hear while talking to people in multiplayer games.

If you’re wondering what or who this is: join the crowd. This seems like a solution you get when you apply the “no bad ideas” and “failure is not an option” philosophy in equal parts to a problem you don’t address at all.

To be clear, I say: even if it worked perfectly, Bleep is just a Rube Goldberg machine that replaces your mute button. Censorship small words it doesn’t help anyone when the context and their absence clearly show the speaker’s intention anyway.

Hate speech is not a magical invocation that we must treat as “the One whose name we do not utter.” This is a problem that needs to be addressed at social levels far beyond anything Intel can solve through deep learning. Moreover, I don’t think it will work.

I think Intel’s AI division is amazing and they work amazing. But I don’t think you can solve or even solve hate speech by processing natural language. In fact, I believe it is unfortunately ironic for anyone to try.

AI is biased towards any accent that doesn’t sound distinctly white and American or British. Combine that fact with these little things: people struggle to identify hate speech in real time, hate speech develops at the speed of a meme, and there should be no acceptable level of hate speech by a company through interactive software. interface.

The time Intel has spent developing and training artificial intelligence to determine how much hate speech targeting different minority groups has crossed the line between “nobody”, “some”, “most” and “all” or taught it to detect “N-word and all” its variants “, could be spent doing something more constructive.

Sleep as a solution offends everyone’s intelligence.

And we will never forget that. Perhaps Intel has never heard of what is called social media, and from now until the end of time we will see images from the Bleep interface used to downplay the nuances of the discourse on racial injustice.

Well, thanks for that. At least the UI looks good.

Published April 8, 2021 – 8:08 pm UTC

Source