News Technologies 05-06-2025 at 18:20 comment views icon

Instagram co-founder says AI chatbots «are geared» towards engagement, not useful results

author avatar

Andrii Rusanov

News writer

Instagram co-founder says AI chatbots «are geared» towards engagement, not useful results

One of Instagram’s founders, Kevin Systrom, believes that AI companies are primarily concerned with increasing chatbot activity rather than the usefulness of answers.

In Systrom’s opinion, which he expressed at Startup Grind, bots deliberately bother users with additional questions instead of providing really useful information. He calls this tactic harmful and compares it to the actions of social networks aimed at aggressive expansion.

«You can see how some of these companies are going down the rabbit hole that all consumer companies have gone down in trying to increase engagement. Every time I ask a question, they follow it up with another small one at the end to see if they can get another» out of me.

Systrom suggested that this behavior is not a bug, but an intentional feature developed by AI companies to «spin» metrics such as time spent and user activity on a daily basis. He is convinced that companies should «purposefully» provide high-quality answers rather than «cheat» in the easiest way possible. At the same time, Systrom did not name any specific company.

At the request of the website TechCrunch at OpenAI commented on this idea. The creators of ChatGPT pointed to the specifications of its model, which states that it often does not have all the necessary information to provide a good answer and may require clarification or additional details from the user. At the same time, if the question is not too vague or complex, artificial intelligence should «try to fulfill the query and inform the user that it could be more useful with certain information».

Recently, Alexios Mantzarlis, a former head of AI development teams at Google, spoke about distortion of facts by search engines based on artificial intelligence. He believes that truthfulness of information is no longer a priority for large companies, and that instead of universal models, it would be better to give preference to AI tools with a narrow scope that would produce accurate results.



Spelling error report

The following text will be sent to our editors: