Framedrop.ai

True journalists don't want to rewrite others' stories. They want to find and break their own. I think about this a lot when I hear objections from journalists who refuse to use AI to write because they want to use their own "vocabulary." However, the majority of the pieces journalists write are not original stories. They are: - Repurposed content from other sites. - Things someone said during an interview. - Summaries of events and other news. What if AI could free up journalists' time to investigate and find stories no one told yet?

The economic impact of gen AI in the media & entertainment industry is expected to be $35-55B 🤯 And this is only for North America 🇺🇸 🇨🇦, according to McKinsey. The biggest part of this impact is expected to come from enhanced productivity of employees, improving creation, production, and management of content. Content should be the priority because: 👀 It's what people look for and engage with. 🤑 It's what media companies monetize. 🧑🏭 It's what takes most of the effort and resources. At Framedrop.ai we believe content repurposing is one of the best ways we can apply AI to improve its quantity without compromising on the quality. - Converting TV feeds and news into web articles. - Converting documentaries and videos into shorts. - Enrich existing content with captions, b-rolls, etc, and distribute it directly on other social media channels. We see these use cases having a huge impact on our clients' productivity already. It's the same quality content humans have produced but repurposed by AI. A really nice example of how humans and AI can work together 😊

We see this happening every day: Competitors of our clients trying to keep up with their content production speed and failing. In media, especially in the news space, if you're not fast publishing about the latest events and hot topics, you miss the wave very quickly. 𝗢𝘂𝗿 𝗰𝗹𝗶𝗲𝗻𝘁𝘀: 5 daily short clips about what's happening on that same day. 𝗧𝗵𝗲𝗶𝗿 𝗰𝗼𝗺𝗽𝗲𝘁𝗶𝘁𝗼𝗿𝘀: 5 clips a week, outdated and irrelevant by the time they are published. The tools already exist, but those not using them will be left behind.

"A paywall tells readers that there’s something behind it that’s worth paying for." This is an excerpt from Sara Morrison's prediction for journalism in 2025. She makes a brilliant point about the fact that quality content has a cost and that, amid too much low-quality AI-generated content, people will prefer to pay for it. In her own words: "Free content will be increasingly synonymous with AI slop and other low-quality content that’s barely worth anyone’s time to read, let alone pay to advertise on." Her claim is that paywalls will increase in popularity and that quality content will be easier to monetize thanks to the flood of AI content no one wants to read. 😅 I agree with it which is why we are focused on helping media outlets repurpose this quality content using AI in order to increase its distribution and traffic to their paywalls. In this case, AI is not used to generate novel content but to repurpose quality content into different formats.

We're producing 10x longer and better articles for our clients compared to human-written ones. I'll take one specific client as an example to illustrate this. They have a TV show for which they produce web articles with short clips with the highlights. The articles showcase the show and attract people to the website. However... Their website's average article length: 1 sentence 😬 When you're talking about 10-20 articles per day, it's impossible for the team to write more than 1 sentence. Now, with @Framedrop.ai, their articles have on average 10 sentences. This is great: - For SEO, the text will include more keywords and places for internal linking to other articles. - For grabbing people's attention, making them excited about watching the video (which may or may not have ads). We're still testing these but so far so good! PS: we also plan to make articles even longer to increase the value they provide (10 sentences is still too short).

Social media is winning. Traditional TV is struggling to keep up. Gen Z’s attention span? 1.3 seconds. 😱 They don’t sit through long broadcasts. They expect fast, engaging, and personalized content. (𝘢𝘤𝘤𝘰𝘳𝘥𝘪𝘯𝘨 𝘵𝘰 𝘢 𝘨𝘭𝘰𝘣𝘢𝘭 𝘴𝘵𝘶𝘥𝘺 𝘣𝘺 𝘠𝘢𝘩𝘰𝘰 𝘢𝘯𝘥 𝘖𝘔𝘋 𝘞𝘰𝘳𝘭𝘥𝘸𝘪𝘥𝘦) In the US, social media is surpassing TV, and the gap is only growing. ⚠️ The problem? Most media companies are still using slow, manual workflows to create short content. By the time it’s published, the moment has passed. Rushing to keep up? That often means sacrificing quality. 🚀 The solution? • Short clips, published fast. • AI-generated articles, tailored to your audience. • Content where Gen Z actually consumes it. The job is not harder now, it's just different. Know someone in media who needs to hear this? Share this video with them.

This is what media leaders believe AI will be most important for in 2025 📺 ✨ For me, this is one of the most interesting charts from the Reuters Institute for the Study of Journalism 2025 predictions report. Not only because it's about AI, about the future, but mostly because it tells us how media leaders see the role of AI in the media outlets. Two interesting findings from it: -> 𝟵𝟲% of the surveyed leaders see AI as something very or somewhat important for 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗻𝗴 𝗯𝗮𝗰𝗸𝗲𝗻𝗱 𝘁𝗮𝘀𝗸𝘀 such as transcription, copyediting, etc. These are repetitive tasks that obey to patterns and defined rules, perfect for AI and boring for humans. -> 𝟳𝟳% of the surveyed leaders see AI as something very or somewhat important for 𝗰𝗼𝗻𝘁𝗲𝗻𝘁 𝗰𝗿𝗲𝗮𝘁𝗶𝗼𝗻 WITH human oversight. We're talking about generating summaries, titles, headlines, metadata, and even graphics and video. These numbers are incredibly positive and reassuring. The media industry is embracing AI as a one-in-a-lifetime opportunity to do more with less. I'm super excited to help more and more media outlets over the next years. Let's do this 💪

AI is the new status quo in media. No way around it. But there are challenges: • Resistance to change • Overloaded journalists • Complex production processes One clear solution? 𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴. Media professionals need the right tools to make their workflows faster and more efficient. At Framedrop.ai, we’re thinking about how we can contribute, e.g. webinars, blog articles, workshops? I’d love to hear your ideas! Drop them in the comments. And congrats to João, Mariana and the team on launching the Digital Media Hub at AMCC - Aveiro Media Competence Center, a big step for the industry! 🚀 Partners: Google NOS SGPS JCDecaux Associação Portuguesa de Imprensa Universidade de Aveiro Lusa Cenjor - Centro Protocolar de Formação Profissional para Jornalistas

64% of media and entertainment companies have been using generative AI for ~1 year 💡 This data comes from a survey Google Cloud and NRG did on 263 senior leaders in global media companies. On top of the great early adoption, media companies are seeing incredible business growth attributed to their gen AI initiatives: - 66% have experienced increased conversion rates. - 63% have increased leads and customer acquisition. - 65% have successfully created new products/services. Most of this is due to streamlined operations, empowering employees, and personalized experiences. And no, no one is getting fired! Quite the opposite. -> 34% of these leaders report that employee productivity has at least doubled! 🤯 In a few years, the media industry will be a lot more efficient and productive thanks to AI. This is only the start.

Which one of these articles is generated by AI? 😉 Let's play a little game I like to call "AI or Human?": 𝗕𝗮𝗰𝗸𝗴𝗿𝗼𝘂𝗻𝗱 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 Today has a show called 'Jenna and Friends' where Scarlett Johansson made an appearance recently. As with any other TV show, they are turned into highlight mini-videos and articles and publish on their website (usually done by humans by hand!). The two articles below come from the same episode of the show but: - One was generated by a writer from their team. - The other was generated by Framedrop.ai. Which one is human-written? 😅