Jordan Scheltgen ⚫️🟡’s Post

I am fed up with people blaming #algorithms as if they are not human actions. Algorithms are editors. Like any editor they exist to sell the #media they determine. Editors who lie in order to sell newspapers lose their jobs. People who create algorithms to maximise profits from advertising regardless of the nature of the content should be regulated the same way. The current disturbances are partly a result of this failure of regulation. It is not acceptable simply to blame the algorithm. The algorithm must cease to be a protection from responsibility.

Brit Morin, Jessica E. Lessin, Sam Lessin and Dave Morin had a really interesting conversation on a podcast this week about media businesses, the tribulations of the The Washington Post and how it's losing piles of money and subs. The discussion dipped into something Chris Anderson coined 20 years ago, perhaps before their time, called "long tail" content. Essentially his thesis called on creators to hyper-focus only on their niche, becoming the expert in hacking heat pumps or the best practices of dog grooming. Now, I'm one of those always looking to appeal to the broadest audience possible so naturally I've pushed against this framework, considering it too limiting. Then I met and worked under Robert Allbritton at POLITICO, where in spite of many in the company not on board with the philosophy, he talked about starting the company to deliver value to a sliver of the audience: those most powerful in the halls of government and soup of politics (and charge significant money for it). Needless to say it was a winning approach, because it's unlikely Axel Springer paid a billion dollars a few years ago for the ad-supported general news side of the house - as terrific as that content is. It was the subscription "Pro" product that overdelivered to that wonky sliver. And while the chatters on the pod did not say so, a similar strategy could well work for the Post: Turn 80% of the operation toward finding a little white space in the "political and policy makers" space, and become an absolute authority there ... and then, of course, license all that to the LLMs ;-) (But that's a story for another post.)

I'm really excited about this one. How to create a client-ready first draft in minutes, rather than hours. Featuring.... Ivy... MVPR's platform brain 🧠. Ivy is trained on over a year of interactions with journalists, and has the power of our Journalist Vector Search behind it. Ivy knows what makes a great piece of content, and allows you to add the information in as you gather it. Creating a first draft now takes minutes rather than hours. And if that wasn't enough, Ivy is then able to suggest journalists that are specifically relevant to your content, based on our semantic analysis of what the media is writing about every day. It's AI-enabled workflows like this one that are allowing us to do in 15hrs what a traditional agency achieves in 80hrs. Please note, the announcement in this video is not real - but Ivy and the power it has, is. And if anyone is wondering how long the raw version of this video was before I tried to shorten it to make it faster to watch: it took me 4 minutes and 5 seconds to record this demo.

Balancing Free Speech and Harmful Content Online 🌐 www.enochchan01.com 📰 My Blog: https://lnkd.in/gDEjyfX6 👥 fb connect: https://lnkd.in/gvyifNHc 📺 Subscribe to my YouTube channel: https://lnkd.in/gg2_xKBM Finding the right balance between protecting free speech and mitigating the spread of harmful content online is one of the most pressing challenges of our time. It requires a nuanced approach that respects fundamental rights while also acknowledging the potential for online platforms to be used to incite violence, spread misinformation, and cause real-world harm. What strategies do you think are most effective in achieving this balance? Let's discuss the complexities of content moderation, the role of technology, and the importance of international cooperation in tackling this issue. Share your thoughts and insights below. #freespeech #onlinesafety #contentmoderation #digitalcitizenship #internetgovernance #harmfulcontent #misinformation #technologyethics #onlineethics #socialmediamarketing #digitalwellbeing

Thanks for the reference Benjamin R. - to develop this discussion, I think the current challenge is what a half-way point to agentic workflows looks like for companies, and what the human-machine interface looks like before workflows are fully agentic or agent-assisted. Most of us can imagine what a fully agentic workflow will look like: request in -> completed task out. It's much harder to breakdown the process involved in completing a task and then decide whether each incremental step in the process is better done by a human or the machine. That requires deep sector-specific expertise and an understanding on what data science, AI and automation is capable of delivering. If the task is better completed by a human, what should that interface look like? If it's a machine, what contextual information does it need to work to the highest standard? And then how do you pass the baton from one to the other to complete a complex process? For most companies, especially services companies, I expect the next phase of genAI development will not be how they can train staff to use next-word-predictors, but how to design or adopt an interface that stores the right knowledge while it works to complete processes.

One thing most professionals struggle with in the age of LLMs is this—they can’t be weird or quirky with prompts Feeding LLMs the same playbook to act as content strategists/CMO's just churns out identical results. It’s like those modern buildings that look sleek but have zero personality—functional, sure, but uninspiring. Mediocrity may be "good enough" for most, but it’s still mediocrity. Creating insane, standout content requires more than just a formula. It takes bold taste, unapologetic quirkiness, and strong opinions. And yeah, be weird. It helps!

Entrusting your money to a financial institution is a vulnerable decision. You can help put people at ease by establishing yourself as experts who will keep their money safe. How can you achieve that? Publishing content! Here are the top 3 benefits of content marketing for financial institutions.

One of the enduring lessons that has informed the strategy of The Air Current is an existential need to 'own your own platform'. That's editorial freedom to control your own destiny. If you live by the algorithm you'll die by it, too. (It will decide if you see this post at all, in fact.) So our growth plan has always been to just go around it. Your subscription is deeply appreciated and makes all our reporting possible, but just sharing your email address (which we never sell) so we can reach you directly is incredibly important, too. Please consider adding your name to TAC's distribution list to keep on top of our latest scoops and deep reporting.

This short got 10,000,000 views in 24 hours. A channel record Why? POWER LAWS 📈 Basically 10% better hook or editing or retention doesn’t mean 10% more views... …It can mean 10 TIMES MORE VIEWS For this short: - Great retention - 10% over average - Broad idea - tap into biggest audience - Perfect hook We wouldn’t even have the time to figure out high leverage things like this if we were spending all of our time editing content. Thanks to my http://clipt.co editors for breaking this new record! Check out the short (I bet you’ve wondered this exact thing)👇

Our decentralised, open-source — and free-to-media — newswire will be launching very soon. Via the Newspage newswire, everyone decides what's written, everyone helps write it and everyone is an editor — denoising, verifying, fact-checking and ensuring our standards are met. No single person or entity having editorial control is the only way to achieve news without narratives. This is news as it should be (and needs to be to survive): decentralised, democratic and bottom-up, not centralised, elitist and top-down.