The AI Meeting Paradox
Posted on: 2026-03-30
Today I want to talk about how AI is transforming meetings. I have noticed more and more people using it during discussions. In the past, it was already common to have laptops open in meetings, and sometimes a few people would search Google while someone was speaking. Today it feels different. Many participants are actively using tools like Gemini or ChatGPT while the meeting is happening.
When someone speaks, there is often someone else in the meeting querying an AI system about what was just said. In some ways this can be useful because it provides quick validation or additional context. However, it also introduces a new form of confirmation bias. If someone disagrees with a point, it is very easy to ask AI for counterarguments. The result is that a discussion between people can quickly turn into a debate where participants defend their position using AI-generated responses.
In a room where many people are doing this simultaneously, the conversation can become strange. Someone might present a thoughtful argument based on preparation or experience, and another person can immediately produce an AI-generated counterpoint. The problem is that these responses are often only partially understood. People rarely read the entire answer carefully during the meeting. Instead, they quickly scan for statements that support their position.
This can create inefficient discussions where people go back and forth with shallow arguments generated in real time. Someone who prepared carefully might see their argument challenged by a quick AI prompt that produces a counterpoint which is not deeply analyzed. At the same time, another participant who is also using AI might generate yet another response to defend the original position. The conversation can drift away from thoughtful reasoning toward a rapid exchange of AI-assisted talking points.
AI is also changing what happens after meetings. Many meetings are now automatically recorded, which was not always common. In the past, even when meetings were recorded, very few people would watch the entire recording afterward. Today AI can summarize the meeting almost instantly. Because of that, some participants are barely listening during the meeting itself. They attend, but rely on the idea that they can read the AI summary later. AI can even highlight specific parts of the conversation and identify what each participant said. This creates a new dynamic. Statements made during meetings can be easily traced back and analyzed later. If someone makes a claim that turns out to be inaccurate, it becomes much easier to point to the exact moment where it was said. In some cases, people could even search through many past meetings to find examples where a statement was incorrect.
The long-term consequence of this may be that people become more cautious in meetings. Knowing that everything is recorded, searchable, and summarized by AI could change how freely people speak or speculate during discussions. AI is therefore transforming meetings not only during the conversation itself, but also in how those conversations are remembered and revisited afterward.
I will conclude that AI in meetings will likely go one step further with the analysis of how efficient, accurate, and useful someone’s contributions are. These signals will probably emerge at some point. Another transformation is how little preparation people may bring when they rely on AI to generate documents, read AI summaries, and construct arguments on the fly. As a result, the skills of preparing, understanding, mastering a topic, speaking clearly, and even forming independent opinions could gradually dilute.