Recently, the field has been greatly impressed and inspired by OpenAI's ChatGPT. It is undoubtedly clever, capable, and very fun to talk to. Its multi-faceted abilities are significantly beyond many NLP researchers' and practitioners' expectations based on the impression of (not-that-strong) original GPT-3. The natural question is how ChatGPT gets there, and where these fantastic abilities come from. In this post, we try to dissect the emergent abilities and trace them to their sources, hoping to give a comprehensive roadmap about how the GPT-3.5 model family, along with related large language models, evolved to their current forms.

The Daily is made by Lisa Tobin, Rachel Quester, Lynsea Garrison, Clare Toeniskoetter, Paige Cowett, Michael Simon Johnson, Brad Fisher, Chris Wood, Jessica Cheung, Stella Tan, Alexandra Leigh Young, Lisa Chow, Eric Krupke, Marc Georges, Luke Vander Ploeg, M.J. Davis Lin, Dan Powell, Dave Shaw, Sydney Harper, Robert Jimison, Mike Benoist, Liz O. Baylen, Asthaa Chaturvedi, Rachelle Bonja, Diana Nguyen, Marion Lozano, Corey Schreppel, Anita Badejo, Rob Szypko, Elisheba Ittoop, Chelsea Daniel, Mooj Zadie, Patricia Willens, Rowan Niemisto, Jody Becker, Rikki Novetsky, John Ketchum, Nina Feldman, Will Reid, Carlos Prieto, Sofia Milan, Ben Calhoun and Susan Lee. Our theme music is by Jim Brunberg and Ben Landsverk of Wonderly.

What's happened: Artists are now able to opt out of the next version of one of the world's most popular text-to-image AI generators, Stable Diffusion, the company behind it announced. Creators can search a website called HaveIBeenTrained for their works in the data set that was used to train Stable Diffusion, and select which works they want to exclude from the training data. Why it's important: The decision comes amid a heated public debate between artists and tech companies over how text-to-image AI models should be trained. The artist couple who created the website hope that the opt-out service will temporarily compensate for the absence of legislation governing the sector. For the past five years or so, barely a week has gone by without a study, comment, or press release about the potential benefits of psychedelic drugs.

War in Europe, a momentous volcanic eruption and a surprise finding that could rewrite our understanding of reality – 2022 really has been a busy year for science, technology, health and environment news, and all that happened in just the first few months. From stunning space imagery to pig heart transplants, here are the New Scientist news editors' picks of the biggest scientific developments, discoveries and events of the year. Russia's invasion of Ukraine in February has sparked devastation across the country and affected many areas of life around the world, as both nations play a key role in the global supply chains for energy, food and more. It has also raised the spectre of nuclear weapons, with Russian president Vladimir Putin making not-so veiled threats about deploying his atomic arsenal. Thankfully, Armageddon has been avoided, but Russia's offensive has sparked discussion of a new kind of nuclear war, as Ukraine's nuclear power plants became a battleground this year.

I think there's an important set of lessons for us about what the next decade's going to be like for AI. The first is where it came from, which is a team of three people poking at an idea in, like, a random corner of the OpenAI building. This one single idea about diffusion models, just a little breakthrough in algorithms, took us from making something that's not very good to something that can have a huge impact on the world. Another thing that's interesting is that this was the first AI that everyone used, and there's a few reasons why that is. But one is that it creates, like, full finished products. If you're using Copilot, our code generation AI, it has to have a lot of help from you.

As they tinkered with the model, everyone involved realized this was something special. "It was very clear that this was it--this was the product," says Altman. We never even had a meeting about it." But nobody--not Altman, not the DALL-E team--could have predicted just how big a splash this product was going to make. "This is the first AI technology that has caught fire with regular people," says Altman. DALL-E 2 dropped in April 2022. In May, Google announced (but did not release) two text-to-image models of its own, Imagen and Parti. Then came Midjourney, a text-to-image model made for artists. And August brought Stable Diffusion, an open-source model that the UK-based startup Stability AI has released to the public for free. The doors were off their hinges. OpenAI signed up a million users in just 2.5 months. More than a million people started using Stable Diffusion via its paid-for service Dream Studio in less than half that time; many more used Stable Diffusion through third-party apps or installed the free version on their own computers. And then in October we had Round Two: a spate of text-to-video models from Google, Meta, and others. Instead of just generating still images, these can create short video clips, animations, and 3D pictures. The pace of development has been breathtaking. In just a few months, the technology has inspired hundreds of newspaper headlines and magazine covers, filled social media with memes, kicked a hype machine into overdrive--and set off an intense backlash. This story is part of our upcoming 10 Breakthrough Technologies 2023 series. Sign up for The Download to get the full list in January. "The shock and awe of this technology is amazing--and it's fun, it's what new technology should be," says Mike Cook, an AI researcher at King's College London who studies computational creativity. "But it's moved so fast that your initial impressions are being updated before you even get used to the idea.

We predicted that tensions would flare between the U.S. and China over AI in 2022. This proved all ... [ ] too true. As we do every year, last December we published a list of 10 predictions about the world of artificial intelligence in 2022. To keep ourselves honest, with 2022 now coming to a close, let's revisit these predictions to see how things actually played out. There is much to learn from these retrospectives about the state and trajectory of AI today.

The AI research team OpenAl has released a beta version of ChatGPT, a chatbot built on the GPT-3.5 language model. ChatGPT is a chatbot that uses deep learning to generate text that mimics that of a human. It was developed by the firm so that users could get responses that were both technical and plain-spoken. Similar to a personal tutor who is versed in all disciplines, it may react to a variety of inquiries in a natural way. As a result, it is promoted as a Google substitute.

On Wednesday, a designer and engineer named Arvind Sanjeev revealed his process for creating Ghostwriter, a one-of-a-kind repurposed Brother typewriter that uses AI to chat with a person typing on the keyboard. The "ghost" inside the machine comes from OpenAI's GPT-3, a large language model that powers ChatGPT. The effect resembles a phantom conversing through the machine. To create Ghostwriter, Sanjeev took apart an electric Brother AX-325 typewriter from the 1990s and reverse-engineered its keyboard signals, then fed them through an Arduino, a low-cost microcontroller that is popular with hobbyists. The Arduino then sends signals to a Raspberry Pi that acts as a network interface to OpenAI's GPT-3 API.

Check out all the on-demand sessions from the Intelligent Security Summit here. One of the biggest topics underlying the hype bonanza since OpenAI's release of ChatGPT two weeks ago has been: What does this mean for Google search? But it was only on Tuesday evening that Google appeared to finally weigh in on the topic: CNBC reported that employees raised concerns at a recent all-hands meeting that the company was losing its competitive edge in artificial intelligence (AI) given ChatGPT's quick rise. "Is this a missed opportunity for Google, considering we've had Lamda for a while?" read one top-rated question. Alphabet CEO Sundar Pichai and Jeff Dean, the long-time head of Google's AI division, responded to the question by saying that the company has similar capabilities in its LaMDA model, but that Google has more "reputational risk" in providing wrong information and therefore is moving "more conservatively than a small startup."