Five practitioners with wildly different backgrounds and perspectives share their hopes and fears for the future.
It's hard to know where to start with the furious pace of development in artificial intelligence technology. Is it simply a cheap gimmick and a passing stock market fad? Or is it a giant leap toward creating conscious machines and a potential risk to human life on the level of pandemics and nuclear war? |
Several months ago on a frosty February night — indulging in an occasional bad habit of watching music videos too late into the night before heading to bed — I happened across an A.I.-generated video that arrested my attention. While it feels comfortable to dismiss much A.I.-generated imagery as tacky, uninspired and full of errors, this video immediately jumped out as attempting something different. Instead of using A.I. to paint over an existing music video or to produce something in a desired artistic style, it leaned hard into the aesthetic qualities of the machine generation itself — constantly shifting and searching for visual possibilities as it transformed from one frame to the next. |
Watching the mutating characters and scenes, I felt like I was being shown something profound about the nature of the deep learning models that power current-generation A.I. technology: how they grasp for familiar patterns to structure randomness into something approaching coherence, how they ingest the cultural output of humanity and regurgitate it in unfamiliar combinations, and how they can rapidly generate a thousand slightly different variations on a theme — like when you ask ChatGPT the same question over and over again and get another coherent-but-different response each time. |
 | | Generated by Stable Diffusion, with Sagans |
|
I knew that I wanted to find a way to work with Sagans, the European art collective that had used A.I. to craft the music video's visuals. I also knew that I wanted to hear more from professionals who had spent solid chunks of their careers working with the technology. There have been many stories in the news lately of the budding genre: "A journalist plays with A.I. for the first time and is surprised." It's true that our first encounters are part of the story. But what do those with extensive hands-on experience like to gossip about around their lunch tables? What keeps them up at night? |
Working with Phoebe Lett, an audio producer at Times Opinion, we designed an exercise to fit. Phoebe recruited five practitioners with wildly different backgrounds and perspectives: an industry pioneer, an artist, an ethicist, a writer and a concerned computer scientist. We asked them the same set of eight big questions about the nature and future of A.I. Then we passed their answers along to Sagans to be illustrated by their illuminating style of machine animation. |
As you listen to the practitioners agree and disagree with each other about where this is all headed, and watch the animations flicker between what feels like an infinite number of potential pictures of the discussion, I hope you can internalize a fresh sense of the state of the debate and the deeper nature of the technology that is poised to upend so much about our culture and economy over the months and years to come. |
Here's what we're focusing on today: |
Forward this newsletter to friends to share ideas and perspectives that will help inform their lives. They can sign up here. Do you have feedback? Email us at opiniontoday@nytimes.com. |
|
No comments:
Post a Comment