Did you know you might not own the DeepSeek model output you get??! Let’s talk AI terms of use and how two major platforms, DeepSeek and OpenAI, handle your data and rights differently. As a product manager, I think it’s critical to understand what you’re agreeing to when you use these tools—because some terms might surprise you. Here’s a quick breakdown: • Privacy: DeepSeek monitors user inputs/outputs heavily and keeps data even after account deletion. OpenAI, on the other hand, lets you opt out of data being used for training and is less invasive overall. • Ownership: OpenAI makes it simple—you own your Inputs and Outputs. DeepSeek? Not so much. They retain a lot of control and don’t clearly outline your rights over generated content. • Dispute Resolution: OpenAI uses arbitration under California law, giving you a fairer process. DeepSeek? All disputes go through () courts, which could be a nightmare for non-()users. • Monitoring: DeepSeek monitors everything, including databases for “illegal content features.” OpenAI focuses more on keeping its services safe without being overly invasive. • User Control: OpenAI allows opt-outs and provides clear notices for changes to terms or prices. DeepSeek offers far less user control, keeping the terms skewed in their favor. These differences matter, especially if you’re building with or relying on these platforms. Read the fine print—it’s the key to safeguarding your privacy and protecting your rights. What are your thoughts? Drop them in the comments! Hashtags: #product #productmanager #productmanagement #startup #business #openai #llm #ai #microsoft #google #gemini #anthropic #claude #llama #meta #nvidia #career #careeradvice #mentor #mentorship #mentortiktok #mentortok #careertok #job #jobadvice #future #2024 #story #news #dev #coding #code #engineering #engineer #coder #sales #cs #marketing #agent #work #workflow #smart #thinking #strategy #cool #real #jobtips #hack #hacks #tip #tips #tech #techtok #techtiktok #openaidevday #aiupdates #techtrends #voiceAI #developerlife #privacy #future #2025 #question #truth #deepseek #cursor #replit #pythagora #bolt

I don't think this video is going to do very well, but I need to make it anyway. This is a comparison of privacy and data use between OpenAI and DeepSeek, and I think it's worth knowing. So, let's get into it. There's six sections. Number one, privacy and data use. So DeepSeek retains broad rights to review user inputs and user outputs to monitor compliance, including maintaining databases of quote, illegal content features. Even after account deletion, it reserves the right to retain data. It provides very little clarity on how it de-identifies you when using data. OpenAI allows users to opt out of having their content used to train models. It clearly describes what it's doing with inputs and outputs to improve services. It does not monitor or maintain compliance databases, and it retains user ownership of inputs and outputs very clearly. So if you put something in, you own it at OpenAI. That's not clear at DeepSeek. Number two, ownership and licensing. Yes, we're going farther there. DeepSeek grants a revocable, non-exclusive license for users to use the service and retains most rights on the platform, requiring explicit consent for any other use. It vaguely defines user rights over outputs with potential restrictions on how the outputs the model gives can be used or shared. So like if it comes back with an answer, it's not exactly clear what your legal rights to that answer are. OpenAI provides very clear and user-friendly ownership terms. It allows users to own the inputs and assign ownership of the outputs to users. It's transparent about the potential non-uniqueness of outputs because, of course, large language models recognize patterns. Number three, jurisdiction. DeepSeek applies Chinese law and requires disputes to be resolved in Hangzhou, China, and this definitely creates barriers to Americans who wish to have concerns addressed in the courts. OpenAI applies California law and allows for dispute resolution through arbitration, which DeepSeek does not, and it's a very transparent process for appeals. Number four, monitoring and security. DeepSeek implements monitoring mechanisms, including reviewing user activity, keystrokes, establishing, quote, risk filtering mechanisms, which they don't define. OpenAI just wants to keep the service safe and functional, or that's the way they write it in the terms of service, and encourages users to flag inappropriate behavior, and they give like up and down thumbs to like call that out, and explains the limitations of AI-generated content, inaccuracy risks, probabilistic nature. Those are all important things. Number five, account termination. DeepSeek explicitly states that certain data will be retained, or may be retained, after account deletion, and includes severe penalties for violations, including permanent bans, data retention for legal purposes, and cooperation with Chinese authorities. OpenAI reserves the right to terminate accounts for policy violations, but it provides an appeals process, and it deletes inactive accounts, and provides advance notice. Number six, transparency and user control. DeepSeek retains significant control over the platform, outputs, user data, with limited mechanisms for user feedback and control. OpenAI really outlines what data can be used for, and offers 30 days notice for any price increases. DeepSeek doesn't offer any ability to sort of opt out of data training. All right. I think I've done enough here. At the end of the day, DeepSeek's terms feel really invasive to me, even by the standards of Silicon Valley. I do not have to believe that OpenAI is perfect, and that DeepSeek is evil, which I don't. I think DeepSeek is a great model to be concerned about this. This is something that is worth noting, and yes, I am very aware that DeepSeek can be run locally, and you don't have to apply any of these, but I will tell you, you don't reach number one on the App Store by being run locally. You reach number one on the App Store by people who don't understand what running locally means, downloading the app, not reading these terms of service, and just starting to use it. And that is a very significant stance. That's a significant change. What would happen if DeepSeek, because the ownership of outputs is ambiguous, were to assert ownership over a startup idea that came out of a DeepSeek output? That's not clear. We may have that court case. I don't know. But the point is that this is a concerningly broad terms of service, and I felt like I needed to make it, not because I think OpenAI is the greatest. I don't. Not because I think OpenAI is phenomenally good as a company. Companies are companies. They do what they do. But because I think this is really, really invasive. So there you go. I don't think this video will do well, but I wanted to let everyone know, and if you're watching this, maybe you'll care. Maybe you'll share.