There is one time a virtual assistant named Miss Dewey, a humorous librarian played by Janina Gavankar who assists you with your questions during Microsoft’s first attempt at the search engine. Miss Dewey was launched in 2006, complete with over 600 lines of recorded dialogue. She was ahead of her time in a number of ways, but one particularly overlooked example was captured by information scholar Miriam Sweeney in her 2013 book. Doctoral thesis, where she details the racist and sexist implications of Dewey’s responses. That includes lines like, “Hey, if you can get inside your computer, you can do whatever you want with me.” Or finding “trumpeting” caused a clip of her eating bananas to play, or typing in terms like “slums” led her to perform a rap whose lyrics include precious things like, “No, yellow teeth, great slum mutha-fucker BEEP the steps to this part [ass] BEEP.” Sweeney analyzed the obvious: that Dewey was designed to serve the straight, white male user. After all, blogs at the time praised Dewey .’s flirtation.
Ms. Dewey was fired by Microsoft in 2009, but later critics – myself included – would identify a similar form of prejudice in the way some users interact with virtual assistants like Siri. or Cortana. When Microsoft engineers revealed that they had programmed Cortana to firmly reject sexual queries or advances, Reddit was infuriated. One highly rated post read: “Are these damn people serious?! The whole point of ‘her’ is to do what people tell her to do! Hey, bitch, add this to my calendar… The day Cortana becomes an ‘independent woman’ is the day that software becomes useless.” Criticism of such behavior flourishes, including from Your humble reporter.
Currently, between repulsion against ChatGPT and its ilk, the pendulum has come back strong and we have been warned against sympathize with these. That’s a point I made after LaMDA’s AI failure last year: A bot doesn’t have to be intelligent for us to humanize it, and that fact will be taken advantage of by scammers. I stand by that warning. But some went so far as to suggest that previous criticisms of people abusing their virtual assistants were naive support in retrospect. Maybe the men who kept calling Cortana “bitch” were up to something!
You may be shocked to learn this is not the case. Previous criticisms of the misuse of AI are not only true, but also predict the more dangerous digital landscape we face today. The real reason the critique moved from “humans are too mean to bots” to “humans are too good to them” is that the political economy of AI has changed dramatically and dramatically, and with it. are sales tactics of technology companies. Previously bots were sold to us as a perfect servant, now they will be sold to us as our best friend. But in each case, the pathological response to each generation of bots implicitly required us to personify them. Bot owners always weaponize our best and worst impulses.
A counterintuitive truth about violence is that, while dehumanizing, it actually requires the perpetrator to see you as human. It’s a harsh reality, but everyone, from war criminals to pub criminals, to some extent, thinks their victims are feeling pain. Dehumanization is not about not seeing someone as human, but the desire to consider someone less than human and act accordingly. Thus, to a certain extent, it is the extent to which people mistake their virtual assistants for real people that encourages them to abuse them. It wouldn’t be fun otherwise. That brings us to the present moment.
Previous generation of AI were sold to us as perfect servants—a sophisticated PA or perhaps Majel Barrett’s Starship Enterprise computer. Humble, know-it-all, always ready to serve. The new chatbot search engines carry some of the same associations, but as they grow, they will also be sold to us as our new confidants, even friends. our new. therapist.
