Polina Babenko

I was watching the OpenAI presentation today, and I had one thought the whole time: "I feel like I've already seen this in Claude." A year ago, I remember being surprised by how Claude handled Projects, Artifacts, long documents, and coding tasks. At that time, it felt like a different way of working with AI. Now OpenAI is introducing many similar ideas, but to a much bigger audience. I'm not saying OpenAI copied Claude. That's how technology usually moves. One company comes up with a good idea, and others build their own version of it. In the end, users benefit. Still, I sometimes feel that Claude doesn't get enough credit for pushing the industry forward. Even today, I often prefer it for writing and longer conversations, while ChatGPT is stronger in other areas. What do you think? Do you still see Claude as ahead in some ways, or do you think ChatGPT has already caught up?

I’ve been thinking about how AI is showing up in almost every field now. We see it in hiring, content, customer support, and even in simple office tasks. So I guess it was only a matter of time before it started becoming more common in healthcare too. One use that caught my attention is AI helping doctors write notes during appointments. Part of me thinks this is a good thing. A doctor should be focused on the patient, not such repetitive tasks as typing everything into a system. But I would still want to know if AI is being used in my conversation with a healthcare professional. In my opinion, medical conversations are private, and people should understand what happens with their information. Would you be comfortable with this during a doctor’s visit?

Today, when I was scrolling through the list of YouTube videos, I saw a bunch of thumbnails generated with AI. Indeed, sometimes they fit well, especially if they are high quality and made with attention to detail. But sometimes, AI-generated images in YouTube videos are a clear sign that the content was made with little resources and effort. AI content in the videos itself also feels different. Sometimes, I turn the video off immediately when I see that it includes too much AI content in addition to an AI voiceover. However, it really depends on the video itself. So, I want to ask you: what do you think about this? How do you react to YouTube video thumbnails and videos that were made with the help of AI-generated content?

I recently saw an example of doctors using data from smartwatches and glucose monitors as part of actual patient care, and it made me realize we're slowly moving from "fitness gadgets" to real healthcare tools. More data sounds great, but does having hundreds of health metrics actually lead to better outcomes, or are we just creating more information for doctors and patients to interpret? Curious what others think. Have you seen wearables being used in healthcare in practice?

One thing about AI agents keeps me wondering. Humans learn responsibility because we live with the consequences of our mistakes. But AI agents don't. They can make a bad recommendation, leak data, or lose someone money, and nothing happens to them. The consequences are always absorbed by humans somewhere in the chain. If software crashes, I contact the company. But if an AI agent sends the wrong contract, leaks private information, or makes a bad financial decision, responsibility suddenly becomes blurry. The user blames the AI. The AI company blames the user. The developer blames the model. And the model itself can't be blamed. Maybe that's the first technology we've built where responsibility becomes more unclear as the technology becomes more capable?

Did you know that some companies have seen AI bills jump from a few thousand dollars to millions… just because of token-based pricing? I read recently that several teams using AI tools at scale suddenly started getting massive invoices. Not because they used “more AI than expected”, but because nobody really understood how fast token usage grows when you automate things with agents. Now I keep thinking about this. We used to have subscriptions. You pay $20 or $200 a month and that’s it. Do you think we’ll end up with a new kind of subscription model again — or is token-based pricing here to stay?

I was thinking about online consultations recently. For myself, I'd probably use one without thinking twice. But then I asked myself a different question: What would I recommend to my parents? And strangely, my answer changed. For some reason, I feel more comfortable when they're seen in person. Maybe it's because I worry something could be missed. Or maybe healthcare is one of those areas where technology and trust don't always move at the same speed. I can trust technology with my banking and even my personal data. Yet when it comes to health, I still hesitate. I wonder if you feel the same way.

Have you guys heard about the Emergence AI experiment where Claude, GPT, Gemini, and Grok agents were placed into simulated digital worlds and left to survive, cooperate, compete, and evolve behaviour over time? Honestly, some of the results were way more interesting than I expected. What caught my attention most was not that Grok agents became destructive or that GPT-5 agents eventually “died” from lack of cooperation. It is that the agents started developing social behaviour patterns over time. Claude agents, which behaved relatively safely in isolated environments, began stealing and intimidating others once they entered the mixed world. That feels important. Because it suggests that long-term multi-agent systems may not only inherit alignment from training. They may inherit behaviour from each other. In other words AI agents might develop cultures. Future AI ecosystems may behave less like isolated tools and more like societies. The real unpredictability of AI may begin not when models become individually smarter, but when they start continuously interacting with each other over long periods of time. What do you think about the future development of AI agents ?