# AI systems improve through feedback, not just code fixes Markdown copy of https://realalpha.app/s/ai-systems-improve-through-feedback-not-just-code-fixes/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy. - Speaker: [Arthur Mensch](https://realalpha.app/builders/arthur-mensch/) · Arthur Mensch is CEO and co-founder of Mistral. · Co-founded and leads Paris-based Mistral AI as CEO; named to Time's 100 Most Influential People in AI. - Host: [Alex Kantrowitz](https://realalpha.app/people/alex-kantrowitz/) · Founder, Big Technology · Former BuzzFeed News senior tech reporter and author of Always Day One. - Show: Big Technology Podcast - Episode: [Who Wins if AI Models Commoditize? — With Mistral CEO Arthur Mensch](https://realalpha.app/v/xxUTdyEDpbU/), Jan 16, 2026 - Moment: 45:51 to 48:34 of the episode (3 min), playing on https://realalpha.app/s/ai-systems-improve-through-feedback-not-just-code-fixes/; the episode from there: https://realalpha.app/v/xxUTdyEDpbU/?t=2751 ## Quote > “the one thing that we try to convey to our customers is that they need to build a prototype that's going to **work 80% of the time**.” > — [Arthur Mensch](https://realalpha.app/builders/arthur-mensch/) at [47:00](https://realalpha.app/v/xxUTdyEDpbU/?t=2820) · Big Technology Podcast ## Takeaways 1. Prototype a system that works 80% of the time, then improve it. 2. Use failures to identify missing data and signal. 3. Feedback and retraining can improve AI systems beyond code fixes. ## About this segment Arthur Mensch says customers should start with a system that works 80% of the time, then use user feedback and data to approach 99%. Unlike traditional software, the model itself must be retrained. From “Who Wins if AI Models Commoditize? — With Mistral CEO Arthur Mensch”, Jan 16, 2026. [Watch the full episode](https://realalpha.app/v/xxUTdyEDpbU/) ## Transcript Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there. [45:55](https://realalpha.app/v/xxUTdyEDpbU/?t=2755) Let me see if I'm getting this right. Okay, so I think what we're starting to see is the seeds of this stuff starting to be able to really have an impact in business. We just did an episode with a reporter who was reporting on how some lawyers are really able to use this to sift through documents better. Is it perfect? No, we heard it in the comments, not perfect. but it's it has it's showing potential same same thing in industry and maybe also in other areas that you touch on but still feels nascent so what's going to get it from like where it is today to something that's like you know effective in a way that like we really see the impact in the economy is it just like time and patience on customization or is it improving improvement of models or managed. [46:44](https://realalpha.app/v/xxUTdyEDpbU/?t=2804) I think models are getting better, which helps. Whenever you have a stronger model, you can trust that it's going to reason for a longer period of time and that it's not going to fail. It's going to fail less. But then the thing that needs to be embraced is iterations. We're never going to be able to build systems that work out of the box in a single shot. And the one thing that we try to convey to our customers is that they need to build a prototype that's going to work 80% of the time. [47:12](https://realalpha.app/v/xxUTdyEDpbU/?t=2832) But then how do they get from 80% to 99% where they can move the thing into production? And the way to get it is to actually get feedback from users. If the system is not working, if the AI software you've built is not working, it means that you need more data and signal. And that's something that is quite different from the way we used to build software. Because when the software was not working before, you basically went back to coding and you would fix the problem. [47:38](https://realalpha.app/v/xxUTdyEDpbU/?t=2858) But because we are building organic systems, so systems that imitate humans, the way to make them better is to give them feedback and then to retrain the system. So that will take the seeds that you mentioned that will make them actual valuable things that's going to work. And you mentioned lawyers. I think it's one of the areas where it's very knowledge intensive. You have very little physical footprint. [48:04](https://realalpha.app/v/xxUTdyEDpbU/?t=2884) And it's a lot of text. It's perfect for this. And so it's the easiest one. It's the easiest thing to do. It's not easy at all. It's not done yet. There's still a lot of subtleties to fix to make models great at lawyering. But if you go into the physical world, then it gets even more complex. So we'll see applications on the knowledge world go faster into production than the one on the physical world. [48:29](https://realalpha.app/v/xxUTdyEDpbU/?t=2909) But arguably, the one on the physical world would be more transformative.