{"id":4546,"date":"2025-06-27T13:45:00","date_gmt":"2025-06-27T13:45:00","guid":{"rendered":"https:\/\/sandbox.hbmadvisory.com\/amplify\/bbc-pilots-ai-tools-for-summaries-and-style-checks\/"},"modified":"2025-06-28T21:48:44","modified_gmt":"2025-06-28T21:48:44","slug":"bbc-pilots-ai-tools-for-summaries-and-style-checks","status":"publish","type":"post","link":"https:\/\/sandbox.hbmadvisory.com\/amplify\/bbc-pilots-ai-tools-for-summaries-and-style-checks\/","title":{"rendered":"BBC pilots AI tools for summaries and style checks"},"content":{"rendered":"<p><\/p>\n<div>\n<p>The BBC has launched two public pilots using generative AI to support its newsroom operations, part of a broader push to explore how artificial intelligence can help with news production while maintaining editorial control and transparency.<br \/>\nThe first pilot uses AI to generate bullet-point summaries of longer articles. These &#8220;At a glance&#8221; summaries are designed to give readers \u2013 particularly younger audiences \u2013 a quick and accessible way to grasp the main points of a story. Journalists generate the summary using a standard prompt, then review and edit the result before publication.<br \/>\n\u201cShort, scannable bullet-point summaries have proven popular with readers,\u201d said Rhodri Talfan Davies, the BBC\u2019s executive sponsor of generative AI, who <a href=\"https:\/\/www.bbc.co.uk\/mediacentre\/2025\/articles\/bbc-to-launch-new-generative-ai-pilots-to-support-news-production\" rel=\"nofollow noopener\" target=\"_blank\">announced<\/a> the pilots. \u201cWe\u2019re going to look at whether adding an AI-assisted bullet point summary box on selected articles helps us engage readers and make complex stories more accessible.\u201d<br \/>\nMany publishers in the US and Europe have introduced summaries over the past year, with many reporting increased engagements with the longer articles from which they were drawn \u2013 despite many people expecting the opposite effect.<br \/>\nThe second pilot, BBC Style Assist, is aimed at improving the speed and consistency with which stories from the Local Democracy Reporting Service (LDRS) are published. The LDRS supplies hundreds of local news stories to the BBC every day, but adapting them to BBC house style is time-consuming. The Style Assist tool, developed by the BBC\u2019s research and development team, uses a custom language model trained on thousands of BBC articles to reformat stories quickly and consistently. A senior journalist then reviews each one before it is published.<br \/>\n\u201cIn line with our AI principles, nothing is published without being checked first by a BBC journalist,\u201d said Davies. \u201cThe AI tool has no role in creating the original story \u2013 which has been researched and written by our LDRS partners.\u201d<br \/>\nThe Style Assist trial is being tested by news teams in Wales and the east of England. The BBC said it would assess whether the tool improves efficiency and allows more local stories to be published.<br \/>\nBoth pilots are part of a cautious, transparency-first approach to AI. Any future rollout, said Davies, \u201cwill depend entirely on the results of these tests and ongoing engagement with editorial teams.\u201d<\/p>\n<\/p><\/div>\n","protected":false},"excerpt":{"rendered":"<p>The BBC has launched two public pilots using generative AI to support its newsroom operations, part of a broader push to explore how artificial intelligence can help with news production while maintaining editorial control and transparency. The first pilot uses AI to generate bullet-point summaries of longer articles. These &#8220;At a glance&#8221; summaries are designed<\/p>\n","protected":false},"author":1,"featured_media":4547,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[118],"tags":[],"class_list":{"0":"post-4546","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-publishing-news"},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/posts\/4546","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/comments?post=4546"}],"version-history":[{"count":1,"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/posts\/4546\/revisions"}],"predecessor-version":[{"id":4548,"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/posts\/4546\/revisions\/4548"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/media\/4547"}],"wp:attachment":[{"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/media?parent=4546"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/categories?post=4546"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sandbox.hbmadvisory.com\/amplify\/wp-json\/wp\/v2\/tags?post=4546"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}