Bravo @ElonMusk for wanting Twitter to become the world’s “most accurate source of information”
Here are 5 things you might want to consider
Bravo for wanting Twitter to become “by far the most accurate source of information about the world”
As you probably know, you’ve got your work cut out for you. If you don’t get a handle on the sea of misinformation, Twitter will drown in it. Advertisers will flee. Both the left and the right will hate you. Customers will leave. You can’t sell, or fold, or you will look like a fool. Your only choice is to carry on, and deliver on what you promised. And the early reviews on your first major challenge weren’t great.
Here are 5 things you might want to consider:
Misinformation travels faster on Twitter right now than elsewhere—about 8 times faster than on Facebook—in part because on Twitter there is less friction in amplifying it.
The problem is about to get much, much worse. Knockoffs of GPT-3 are getting cheaper and cheaper, which means that the cost of generating misinformation is going to zero, and the quantity of misinformation is going to rise—probably exponentially.
As the pace of machine-generated misinformation picks up, Twitter’s existing effort, Community Notes (aka Birdwatch), which is mostly done manually, by humans, is going to get left in the dust.
You are going to need AI. But not (just) the popular stuff. No matter how much data you can collect, don’t count on deep learning. Large language models are without equal at generating misinformation, but they suck at detecting it. Accuracy depends on some reference standard of truth, and they have no way of reasoning about truth. Scaling won’t solve this.
You’re going to need something different. Let’s talk.
— Gary Marcus