Last year I shared DALL·E, an amazing model by OpenAI capable of generating images from a text input with incredible results. Now is time for his big brother, DALL·E 2. And you won't believe the progress in a single year! DALL·E 2 is not only better at generating photorealistic images from text. The results are four times the resolution! As if it wasn't already impressive enough, the recent model learned a new skill; image inpainting.

New technology that blends language and images could serve graphic artists -- and speed disinformation campaigns. SAN FRANCISCO -- At OpenAI, one of the world's most ambitious artificial intelligence labs, researchers are building technology that lets you create digital images simply by describing what you want to see. They call it DALL-E in a nod to both "WALL-E," the 2008 animated movie about an autonomous robot, and Salvador Dalí, the surrealist painter. OpenAI, backed by a billion dollars in funding from Microsoft, is not yet sharing the technology with the general public. But on a recent afternoon, Alex Nichol, one of the researchers behind the system, demonstrated how it works.

Open.AI helps us comprehend how advanced AI systems see and ponder our world. Art is the key to this. By exploring the art that is generated by AI and looking at what captures and inspires us, we can test a broader range of ideas and fixable limitations in order to refine our systems.

OpenAI researchers have created a new system that can produce a full image, including of an astronaut riding a horse, from a simple plain English sentence. Known as DALL·E 2, the second generation of the text to image AI is able to create realistic images and artwork at a higher resolution than its predecessor. The artificial intelligence research group won't be releasing the system to the public. The new version is able to create images from simple text, add objects into existing images, or even provide different points of view on an existing image. Developers imposed restrictions on the scope of the AI to ensure it could not produce hateful, racist or violent images, or be used to spread misinformation.

OpenAI researchers have created a new system that can produce a full image, including of an astronaut riding a horse, from a simple plain English sentence. Known as DALL·E 2, the second generation of the text to image AI is able to create realistic images and artwork at a higher resolution than its predecessor. The artificial intelligence research group won't be releasing the system to the public, but hope to offer it as a plugin for existing image editing apps in the future. The new version is able to create images from simple text, add objects into existing images, or even provide different points of view on an existing image. Developers imposed restrictions on the scope of the AI to ensure it could not produce hateful, racist or violent images, or be used to spread misinformation.

A half decade ago, the world's leading A.I. labs built systems that could identify objects in digital images and even generate images on their own, including flowers, dogs, cars and faces. A few years later, they built systems that could do much the same with written language, summarizing articles, answering questions, generating tweets and even writing blog posts. Now, researchers are combining those technologies to create new forms of A.I. DALL-E is a notable step forward because it juggles both language and images and, in some cases, grasps the relationship between the two. "We can now use multiple, intersecting streams of information to create better and better technology," said Oren Etzioni, chief executive of the Allen Institute for Artificial Intelligence, an artificial intelligence lab in Seattle. The technology is not perfect.

On a sunny morning last December, Iyus Ruswandi, a 35-year-old furniture maker in the village of Gunungguruh, Indonesia, was woken up early by his mother. A technology company was holding some kind of "social assistance giveaway" at the local Islamic elementary school, she said, and she urged him to go. When he got there, representatives of Worldcoin were collecting emails and phone numbers, or aiming a futuristic metal orb at villagers' faces to scan their irises and other biometric data. Two months before Worldcoin appeared in Ruswandi's village, the San Francisco–based company called Tools for Humanity emerged from stealth mode. The company's website described Worldcoin as an Ethereum-based "new, collectively owned global currency that will be distributed fairly to as many people as possible."

The OpenAI consortium has unveiled the next iteration of DALL·E, a multimodal AI that could generate rudimental, low-res images from a text-based prompt. This time around, the system is capable of generating images at higher resolution and with lower latency than the original. DALL·E 2 uses OpenAI's CLIP image recognition system and adds the ability for users to edit the results. They can now select and edit areas of existing images, add or remove elements, mash together two images into a single collage and generate further variations of an existing image. What's more, the output images are 1,024 pixel squares, up from the 256 x 256-pixel canvases generated by the original version.

Recently, researchers from OpenAI published a new paper that proposes a method to address this challenge by creating RL models that know what it means to make progress on a new task, by having experienced making progress on similar tasks in the past. Titled Evolved Policy Gradients(EPG), the OpenAI research paper introduces new meta-learning technique based on the concept of a loss function that qualifies the learning progress. When used in RL models, the EPG method does not encode the knowledge explicitly through memorized behaviors but, instead, it uses an implicitly mechanism through a learned loss function. The EPG end goal is that RL agents that can use this loss function to learn a novel task. In initial tests, EPG seems to improves on standard RL algorithms by allowing the loss function to be adaptive to the environment and agent history, leading to faster learning and the potential for learning without external rewards.

Topic models are some of the most popular ways to represent textual data in an interpret-able manner. Recently, advances in deep generative models, specifically auto-encoding variational Bayes (AEVB), have led to the introduction of unsupervised neural topic models, which leverage deep generative models as opposed to traditional statistics-based topic models. We extend upon these neural topic models by introducing the Label-Indexed Neural Topic Model (LI-NTM), which is, to the extent of our knowledge, the first effective upstream semi-supervised neural topic model. We find that LI-NTM outperforms existing neural topic models in document reconstruction benchmarks, with the most notable results in low labeled data regimes and for data-sets with informative labels; furthermore, our jointly learned classifier outperforms baseline classifiers in ablation studies.