Artificial intelligence is revolutionizing social media - and by 2025, we'll be right in the middle of it. From LinkedIn texts and Instagram videos to election advertising on TikTok - AI-generated content is everywhere and is changing how we communicate, consume and interact. But who really benefits from it? What ethical challenges does this entail? And what does the future of social networks look like?
Christina D'Ilio from netzstrategen in Karlsruhe provides current, exciting insights. She talks about the winners of the AI revolution, the responsible use of AI content and her vision for the future of social media.
Christina D'Ilio in an interview:
Christina D'Ilio
Managing Director | netzstrategen
Ms. D'Ilio, you are responsible for the development of digital strategies at netzstrategen and therefore have many points of contact with the topics of social media and artificial intelligence. What is your view on this?
Probably a little different than you might think. My professional background is not the traditional one of IT or business administration - I have a degree in sociology. As a result, I don't just look at technological progress from an economic or technical perspective, but also with a view to its social impact. I find this approach particularly exciting when it comes to the development and use of artificial intelligence.
Artificial intelligence is no longer a dream of the future, but has been with us - more or less - in our everyday lives for several years now. Do you remember the first time you used AI for LinkedIn and the like?
Yes, I remember that very well. It must have been around the turn of the year 2022/23 when I tried to create a post for LinkedIn with AI for the first time. I was surprised at how good the first result I got from AI (ChatGPT 3.5 at the time) was.
Rijul Gupta, founder of DeepMedia, argues that in a few years almost everything we see on social media will be artificially generated. How do you feel about this?
I think this thesis is realistic. However, in my view, the decisive factor is not whether content is generated artificially or by humans - but whether it is of high quality. A good article informs, entertains or moves me - regardless of whether it comes from a human or an AI. The important thing is that it does not contain any untruths and is well written. If these criteria are met, the origin of the content plays a rather subordinate role.
Are there industries or areas that can benefit particularly strongly from AI in social media?
Anyone who creates content for social media can benefit from AI-supported applications. The decisive factor here is not so much the industry, but rather the quantity and diversity of the content produced. The more extensive and diverse the content is, the greater the benefits that can be achieved through the use of AI. AI also offers enormous potential in the B2B sector: a software company that regularly publishes webinars or whitepapers can use AI to automatically generate LinkedIn posts from the most important key takeaways, personalize email newsletters or use chatbots to pre-qualify potential customer inquiries directly on social media. This allows marketing and sales processes to be designed more efficiently and leads to be targeted more effectively.
How can companies ensure responsible AI use in their social media strategies?
When using AI responsibly, you first need to be aware of the different levels of use. Companies should be particularly careful with both the content they make available to AI and the results it generates.
The following applies to input: everything that is intended for publication (social media, website, catalog, etc.) can also be made available to an AI. Sooner or later, it will come across this data anyway. Data that is not intended for the public, on the other hand, should never be made available to an AI - especially personal information. Many AI systems reuse training data so that it can - at least theoretically - be accessed by others.
The following applies to the generated content: All claims made by the AI must be critically scrutinized. Does the quote that the AI suggests actually exist? Are the dates and contexts correct? Without specific instructions in the prompts, AI tools can "hallucinate", i.e. provide invented information just to fulfill the order.
Automated content and AI-generated texts: How do you assess the danger of social media losing authenticity?
The danger is definitely there. When I open my LinkedIn feed, I recognize many AI-generated posts. They are often very generic and follow the same linguistic patterns. Phrases like "exciting journey", "fascinating insights" or "in today's fast-paced world..." abound. However, it is not the technology that is to blame for this development, but the way in which we use it. With precise specifications aimed at individualizing the texts, I get very high-quality results that stand out from the crowd. This requires either in-depth knowledge of prompting or specialized tools that make this work easier.
And which tools are currently particularly promising for companies that want to use AI in social media marketing?
LLMs (Large Language Models) such as ChatGPT, Gemini or Claude are particularly suitable for social media. As their name suggests, they are large language models that are particularly good at generating content. Their disadvantage, however, is that they are often cumbersome for recurring tasks - such as creating and editing social media content.
We have developed the beonbrand.ai platform for precisely this purpose. It combines the strengths of LLMs - in our case ChatGPT - with the advantages of classic word processing. What's more, users don't need to be prompt professionals to generate high-quality, brand-specific content. In a special area for brand tonality, you can define how the brand should communicate: Rather approachable or professional? Which terms should be used and which should be avoided? These specifications can then be activated in various templates, for example when creating a LinkedIn post.
How is AI changing community management and interaction with users?
One thing is certain: AI will permanently change the way we interact with brands and personalities. One example is AI-supported chatbots that are trained with the data and style of celebrities, influencers or brands. Platforms such as Instagram and Snapchat are already experimenting with virtual avatars that interact with followers around the clock and provide authentic answers in real time.
What developments do you expect in the coming years? Where is social media heading with the increasing use of AI?
There are various theories, but I think a development towards hyper-personalization is the most likely.
In future, platforms such as Instagram or TikTok could generate content in real time that is precisely tailored to my interests and current mood. Images, videos and stories could be generated by AI and perfectly adapted to my preferences - with the aim of keeping me in the app for as long as possible. If Instagram & Co. succeed in the future in using AI to generate appealing content in real time and in large quantities to keep users in the app, influencers and other content creators could become increasingly dispensable for the platforms.
