. But it is also pressing ahead—it announced this past week that it is rolling out this system to its Skype communications tool, and the mobile versions of its Edge web browser and Bing search engine.
Microsoft and OpenAI now feel that testing their technology on a limited portion of the public—a sort of invite-only beta test—is the best way to assure that it’s safe. The recent questionable responses from Bing—and the need to test this technology widely—flow from how the technology works. So-called “large language models” like OpenAI’s are gigantic neural networks trained on gargantuan amounts of data. One common starting point for such models is what is essentially a download or “scrape” of.
Well, yeah… if we’re going to use them to make work more efficient, they best be accurate.