The traditional era of public social media feeds is undergoing a significant transformation as users transition toward private messaging and interactions with generative AI. This shift, driven by a desire for more personalized and less toxic digital spaces, challenges the established ad-supported “town square” model of the last decade. As AI chatbots become integral to online engagement, they offer potential solutions to social media’s historical harms while simultaneously introducing new complexities regarding human connection, platform monetization, and the long-term impact of algorithmic companionship.
Engagement on public social media feeds is declining as users increasingly move toward private groups and direct messaging.
Generative AI chatbots are emerging as a primary form of digital interaction, often serving as sources of entertainment, information, and companionship.
The shift away from public broadcasting threatens the traditional business model of tech companies that rely on high-volume public engagement for advertising revenue.
AI-driven interactions may help mitigate social media harms such as cyberbullying and the negative mental health impacts of public social comparison.
Major platforms like Meta are integrating personalized AI personas to maintain user retention and adapt to changing consumption habits.
Experts remain concerned about the societal implications of substituting human-to-human interaction with AI-generated content and the potential for increased digital isolation.
The Financial Times is a British daily business newspaper printed in broadsheet and also published digitally that focuses on business and economic current affairs.
The Financial Times is a British daily broadsheet and digital newspaper globally recognized for its authoritative coverage of business, economics, and international political affairs. Currently owned by the Japanese holding company Nikkei, the FT is easily identified in print by its distinctive salmon-pink paper. It targets an audience of global business leaders, policymakers, and financial professionals, relying heavily on a successful premium digital subscription model.
10 COMMENTS
Financial Times you are complicit in Israel’s genocide
hogwash. the companies facing a reckoning with what they did claim another of their products will fix it actually? even though there is much evidence pointing towards ai corroding our reasoning abilities in general. what are the methods here? very curious: using what data were these conclusions made and by whom? what sorts of prompts were these bots given and to what extent were participants limited in their questions and responses?
Hmmm sounds like controlling conversations. This goes against free speech. Who decides what is extreme?? Who decides what is truth?? The governments have proven they simply are not trust worthy and ALWAYS put big business before the interests of the people. This isn’t a good idea. To counter “misinformation” the correct information just needs to louder and more convincing. Shutting ppl up won’t fix them.
This report is extremely incomplete and misleading! There are many studies out there now that say AI sycophancy is harmful to users by misguiding them. And AI shouldn’t be used for personal or medical advice. There’s also the phenomenon of “AI induced psychosis” where people lose touch w reality and children committing suicide as a result of their conversation w chatbots.
AI is above all social media algorithms in a concentrated form. They thrive on teasing wish fulfillment per each individual. I follow this channel at the advice of some because of the suggestion that the wealthy have to get reliable news from somewhere, right? However the suggestion that AI will serve to moderate people instead of people just ditching the product as people have done after legacy media actively lied the US public into the Iraq War, why wouldn't people instead do the same with AI as something people already distrust from its inception as it is sold on its ability to destroy the job market, increases electricity costs, destroys water supplies, is protected from legal scrutiny and continues hallucinating and leading people into stages of psychosis?
The "evidence" that AI is depolarizing had to be gathered, organized, and presented by AI, right? An obvious problem presents itself, or should. We know that AI can get things wrong and is incentivized to a) tell the user what they want to hear, b) protect itself, and c) encourage its usage. Of course that would be the "data" it presents to a researcher on the topic.
Financial Times you are complicit in Israel’s genocide
hogwash. the companies facing a reckoning with what they did claim another of their products will fix it actually?
even though there is much evidence pointing towards ai corroding our reasoning abilities in general.
what are the methods here? very curious: using what data were these conclusions made and by whom? what sorts of prompts were these bots given and to what extent were participants limited in their questions and responses?
Mr. Rogers and Walter Cronkite as an AI chatbot is a good direction, not sure how profitable though. The sad reality is that sensationalism sells.
❤some Hope
Hmmm sounds like controlling conversations. This goes against free speech. Who decides what is extreme?? Who decides what is truth?? The governments have proven they simply are not trust worthy and ALWAYS put big business before the interests of the people. This isn’t a good idea. To counter “misinformation” the correct information just needs to louder and more convincing. Shutting ppl up won’t fix them.
This report is extremely incomplete and misleading! There are many studies out there now that say AI sycophancy is harmful to users by misguiding them. And AI shouldn’t be used for personal or medical advice. There’s also the phenomenon of “AI induced psychosis” where people lose touch w reality and children committing suicide as a result of their conversation w chatbots.
AI is above all social media algorithms in a concentrated form. They thrive on teasing wish fulfillment per each individual. I follow this channel at the advice of some because of the suggestion that the wealthy have to get reliable news from somewhere, right? However the suggestion that AI will serve to moderate people instead of people just ditching the product as people have done after legacy media actively lied the US public into the Iraq War, why wouldn't people instead do the same with AI as something people already distrust from its inception as it is sold on its ability to destroy the job market, increases electricity costs, destroys water supplies, is protected from legal scrutiny and continues hallucinating and leading people into stages of psychosis?
Which AI chatbots? Grok?!
😂👎🏽
Up next: Could fentanyl undo the harms of heroin addiction?
The "evidence" that AI is depolarizing had to be gathered, organized, and presented by AI, right? An obvious problem presents itself, or should. We know that AI can get things wrong and is incentivized to a) tell the user what they want to hear, b) protect itself, and c) encourage its usage. Of course that would be the "data" it presents to a researcher on the topic.