Thread

It's paradoxical, I've been interested in AI since my late teens and it was one of the main motivations to study computer science, and yet, the current state, especially of the discussion, infuriates me. I've tried to understand just why, and here are my main takes:

(1) The way the discussion is fragmented. I think this is very much the fault of character limits in social media, but many people pick one aspect and stick to it. E.g. saying that every new technology (calculators, the newspaper, television) lead to people being concerned.

That's definitely true, but then again, this piece of technology is different from everything that came before. Others focus only on the amazing things it can do, which it can, but then again, it's not about the amazing things but also about the impact and the consequences.

Others say "I don't see why people are concerned about consciousness. Maybe we're also just biological computers." Which is also true, but then again, this touches really far reaching fundamental questions of humanity.

We've grown accustomed to this fragmented form of non-discourse, but it does avoid the question what this new thing really is and what the consequences are.

(2) The friggin' AI marketing. Ever since the Google sold the weird pictures that convolutional neural networks create as "dreams." They realized that AI can generate a lot of buzz, whether it's useful (yet) or not, and just went with it. en.wikipedia.org/wiki/DeepDream

Remember the Google employee who straight up asked a conversational AI whether it is conscious? The worst kind of pseudoscience, but really great for marketing. cajundiscordian.medium.com/is-lamda-senti…

Remember Google's AI assistant that claimed to be able to make phone calls and book appointments to you. I'm not sure if IIRC, but didn't it turned out to be fake? Still great PR! ai.googleblog.com/2018/05/duplex…

When the new AI powered chatGPT came alive, there was all kinds of reports that it was "unhinged" or became "depressed." This is the Eliza effect in its purest form. Again, great PR! arstechnica.com/information-te…

And lastly (3) the MONEY. By now it is pretty obvious that a lot of money can be made with AI. Remember that Microsoft invested $10bn into OpenAI? Did you think that Bing search would ever make headlines again? forbes.com/sites/qai/2023…

If I asked you a year ago, would you have believed that Microsoft would beat Google to adding AI features to their office suite? Already there are dozens of startups scrambling to build products on top of this new technology.

Don't get me wrong, I've nothing against the money, but at the same time we're fed the narrative that this is about mankind and the next step of humanity blahblah. Let's just agree this is a gold mine and move on. abcnews.go.com/Technology/ope…