Making Magic with Artificial Intelligence: The Aftershock to the Great Fires of San Francisco, April 13, 2013: Sergey Brin and Everyprompt

The day’s theme was to explore the possibilities. We learned how the artificial intelligence is being used to assess and predict floods, fight wildfires and forecast floods. The models that were used in this show were called generative AI models. These are the content machines, schooled on massive training sets of data, designed to churn out writings, images, and even computer code that once only humans could hope to produce.

Something weird is happening in the world of AI. The field broke out of an artificial intelligence winter in the early part of the century because of the innovation of deep learning. This approach to AI transformed the field and made many of our applications more useful, powering language translations, search, Uber routing, and just about everything that has “smart” as part of its name. We’ve spent a dozen years in this AI springtime. But in the past year or so there has been a dramatic aftershock to that earthquake as a sudden profusion of mind-bending generative models have appeared.

Stability AI, which offers tools for generating images with few restrictions, held a party of its own in San Francisco last week. It announced $101 million in new funding, valuing the company at a dizzy $1 billion. Tech celebrities including Sergey Brin were at the gathering.

Song works with Everyprompt, a startup that makes it easier for companies to use text generation. He says testing tools that can make images, text, or code has left him with a sense of wonder. He has not used a website or technology in a long time that felt very helpful or magical. I feel like I am using magic because of using generative artificial intelligence.

Yet the AI at the core of ChatGPT is not, in fact, very new. It is a version of an AI model called GPT-3 that generates text based on patterns it digested from huge quantities of text gathered from the web. The model, which is available for programmers to use, has already shown that it can answer questions and generate text in a good way. But getting the service to respond in a particular way required crafting the right prompt to feed into the software.

A CEO of a company. This ability to answer questions about love or cocktail recipes made it so that Artificial Intelligence was so fond of it. A company is exploring ways to use a tool called ChatG PPT to help write technical papers. “We have tested it, and it works great,” she says.

OpenAI has not released full details on how it gave its text generation software a naturalistic new interface, but the company shared some information in a blog post. It says the team fed human-written answers to GPT-3.5 as training data, and then used a form of simulated reward and punishment known as reinforcement learning to push the model to provide better answers to example questions.

Jacob Andreas is an assistant professor at MIT, and he says the system might widen the pool of people who are able to use artificial intelligence. “Here’s a thing being presented to you in a familiar interface that causes you to apply a mental model that you are used to applying to other agents—humans—that you interact with,” he says.

Lauren Goode: OK, but how concerned are you that you might eventually fall for misinformation that’s generated by an AI that has taken another reporter’s job?

Lauren Goode: Welcome to Gadget Lab, the podcast where we explore the latest and greatest gadgets in technology. Join us as we unbox, test, and review the newest devices and apps and provide expert analysis on the latest trends in the tech world. If you are a tech enthusiast or just looking for some buying advice, there is something for you at Gadget Lab, so grab your headphones and join us.

OK. That’s not our usual opening. The first word in the introduction of the new generative artificial intelligence system from OpenAI was that. Let’s start our actual introduction by doing it. I’m Lauren Goode, and I’m a senior writer—senior robot writer?—at WIRED.

Lauren Goode: We are also joined by WIRED senior writer Will Knight, who joins us from Cambridge, Massachusetts, and who has such a delightful accent that we couldn’t help inviting him on. Hi, Will.

The 2023 Revolution: From Codex to Copilots for Everyday, Everyday Life in a World with Generative Artificial Intelligence

The combination of large language models and coding resulted in two of the most powerful AI developments we witnessed in 2022: the introduction of the OpenAI Codex Model—a large AI model that can translate natural language inputs into more than a dozen programming languages—and the launch of GitHub Copilot, a programming assistant based on Codex.

Humans must learn to speak machines’ languages to communicate with them in the past. But now, Codex lets us use natural language to express our intentions, and the machine takes on the responsibility of translating those intentions into code. It’s basically a translator between the human imagination and any piece of software with an API.

The same technology used to create GitHub Copilot could be put to use to create Copilots that can be used for nearly any kind of knowledge work, and allow knowledge workers to spend a lot of time on cognitive tasks.

That is an extraordinary step forward in productivity for developers as a community of knowledge workers are wrestling with extraordinary complexity and unprecedented demand for their talents. The pattern is repeated across other kinds of knowledge work, so it will just be the first step of many that will be taken in 2023.

Our increasingly complicated and information-dense world requires more knowledge work every year, imposing ever-greater demands on those workers in every field and industry. Copilots for Everything could offer a genuine revolution for types of work where productivity gains have been few and far between since the invention of the personal computer and the internet.

Answers to those questions aren’t clear right now. But one thing is. Even as the giants lay off people, open access to these models has kicked off a wet hot summer that is revitalizing the tech sector. The next big paradigm is not the metaverse, it is this new wave of artificial intelligence content engines, and it is here now. There was a huge rush of products in the 80s that moved paper to PC applications. During the 1990’s, you could make a quick fortune by changing your desktop products to online. A decade later, the movement was to mobile. In the 2020s, building with generative artificial intelligence is a big shift. This year thousands of startups will emerge with business plans based on tapping into the APIs of those systems. The cost of churning out generic copy will go to zero. By the end of the decade, video-generation systems will be the dominant feature in Tiktok and other apps. They may not be anywhere as good as the innovative creations of talented human beings, but the robots will quantitatively dominate.