{"id":47510,"date":"2021-08-11T12:27:17","date_gmt":"2021-08-11T09:27:17","guid":{"rendered":"https:\/\/forklog.com\/en\/?p=47510"},"modified":"2025-09-01T21:45:45","modified_gmt":"2025-09-01T18:45:45","slug":"openai-unveils-codex-an-ai-tool-for-automatic-code-generation","status":"publish","type":"post","link":"https:\/\/forklog.com\/en\/openai-unveils-codex-an-ai-tool-for-automatic-code-generation\/","title":{"rendered":"OpenAI unveils Codex, an AI tool for automatic code generation"},"content":{"rendered":"<p>The nonprofit OpenAI released a machine-learning tool Codex that translates natural language into code. It is designed for both hobbyists and professional programmers.<\/p>\n<p>\\n\\n<\/p>\n<blockquote class=\\\"twitter-tweet\\\">\n<p lang=\\\"en\\\" dir=\\\"ltr\\\">OpenAI Codex has a qualitatively new capability \u2014 it can write code with sufficient accuracy that users can direct their computer in natural language.<\/p>\n<p>Codex is now in OpenAI API, for people to build new businesses or integrate with existing applications: <a href=\\\"https:\/\/t.co\/n3U4DJ7fqC\\\">https:\/\/t.co\/n3U4DJ7fqC<\/a> <a href=\\\"https:\/\/t.co\/3ETsyXMQ7T\\\">pic.twitter.com\/3ETsyXMQ7T<\/a><\/p>\n<p>\u2014 OpenAI (@OpenAI) <a href=\\\"https:\/\/twitter.com\/OpenAI\/status\/1425153851095543811?ref_src=twsrc%5Etfw\\\">August 10, 2021<\/a><\/p><\/blockquote>\n<p> <script async=\\\"\\\" src=\\\"https:\/\/platform.twitter.com\/widgets.js\\\" charset=\\\"utf-8\\\"><\/script>\\n\\n<\/p>\n<p>OpenAI demonstrated how Codex creates simple websites, elementary games and interacts with other applications via <span data-descr=\\\"application programming interface\\\" class=\\\"old_tooltip\\\">API<\/span>, using user prompts. The tool is also capable of translating code between different programming languages and tackling data science queries.<\/p>\n<p>\\n\\n<\/p>\n<blockquote class=\\\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\\\">\n<p>\\&#8221;With more than a dozen programming languages at its disposal, Codex can interpret simple commands in natural language and execute them on behalf of the user\\&#8221;, the company said.<\/p>\n<\/blockquote>\n<p>\\n\\n<\/p>\n<p>According to the developers, the tool \\&#8221;works especially well\\&#8221; with Python, but it also supports JavaScript, Go, Perl, PHP, Ruby, Swift, TypeScript and Shell.<\/p>\n<p>\\n\\n<\/p>\n<p>To use Codex, a programmer in a dedicated window enters commands in natural language, for example \\&#8221;create a web page with a side menu and a header at the top.\\&#8221; After confirmation, the tool translates the request into code.<\/p>\n<p>\\n\\n<iframe loading=\\\"lazy\\\" title=\\\"vimeo-player\\\" src=\\\"https:\/\/player.vimeo.com\/video\/583546972\\\" width=\\\"640\\\" height=\\\"360\\\" frameborder=\\\"0\\\" allowfullscreen=\\\"\\\"><\/iframe>\\n\\n<\/p>\n<p>Codex is built on the GPT-3 model. Its training data include both natural language and billions of lines of code from open sources, including data from public GitHub repositories.<\/p>\n<p>\\n\\n<\/p>\n<p>Developers said the model is not flawless and requires some patience. However, the tool could prove an invaluable aid to speed up and broaden access to programming, they say.<\/p>\n<p>\\n\\n<\/p>\n<p>OpenAI used an earlier version of Codex to create the <a href=\"https:\/\/forklog.com\/en\/news\/github-unveils-ai-powered-copilot-code-autocompletion-tool\">Copilot for GitHub<\/a> tool, introduced in July 2021. It is similar to text autocomplete tools, offering suggestions for completing lines of code.<\/p>\n<p>\\n\\n<\/p>\n<p>Codex is a more advanced and flexible system that not only completes code but creates it.<\/p>\n<p>\\n\\n<\/p>\n<p>OpenAI also announced the launch of a closed beta-test of the tool. Applications to participate can be submitted by anyone on the company&#8217;s <a href=\\\"http:\/\/openai.com\/join\\\" target=\\\"_blank\\\" rel=\\\"noreferrer noopener nofollow\\\">official website<\/a>.<\/p>\n<p>\\n\\n<\/p>\n<p>In July, OpenAI released the Python-like<a href=\"https:\/\/forklog.com\/en\/news\/openai-releases-triton-a-python-like-language-for-neural-network-development\"> programming language Triton for developing neural networks<\/a>.<\/p>\n<p>\\n\\n<\/p>\n<p>In June, researchers from the <a href=\"https:\/\/forklog.com\/en\/news\/openai-curbs-bias-and-toxicity-in-gpt-3\">organization<\/a> discovered a way to improve the \\&#8221;behavior\\&#8221; of the GPT-3 language model in relation to ethical, moral and social values.<\/p>\n<p>\\n\\n<\/p>\n<p>In March, OpenAI representatives stated that <a href=\"https:\/\/forklog.com\/en\/news\/openai-gpt-3-language-model-generates-4-5-billion-words-a-day\"> GPT-3 generates 4.5 billion words daily<\/a>.<\/p>\n<p>\\n\\n<\/p>\n<p>Subscribe to ForkLog AI news on Telegram: <a href=\\\"https:\/\/t.me\/forklogAI\\\" target=\\\"_blank\\\" rel=\\\"noreferrer noopener nofollow\\\">ForkLog AI<\/a> \u2014 all the news from the world of AI!<\/p>\n","protected":false},"excerpt":{"rendered":"<p>The nonprofit OpenAI released a machine-learning tool Codex that translates natural language into code. It is designed for both hobbyists and professional programmers.<\/p>\n","protected":false},"author":1,"featured_media":47511,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"select":"1","news_style_id":"1","cryptorium_level":"","_short_excerpt_text":"","creation_source":"","_metatest_mainpost_news_update":false,"footnotes":""},"categories":[3],"tags":[438,1190],"class_list":["post-47510","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news-and-analysis","tag-artificial-intelligence","tag-openai"],"aioseo_notices":[],"amp_enabled":true,"views":"39","promo_type":"1","layout_type":"1","short_excerpt":"","is_update":"","_links":{"self":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/47510","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/comments?post=47510"}],"version-history":[{"count":1,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/47510\/revisions"}],"predecessor-version":[{"id":47512,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/posts\/47510\/revisions\/47512"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/media\/47511"}],"wp:attachment":[{"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/media?parent=47510"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/categories?post=47510"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/forklog.com\/en\/wp-json\/wp\/v2\/tags?post=47510"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}