Understanding Proprietary Data in Artificial Intelligence

Over the past several months, our team has broken a string of stories about AI agents — OpenAI's hiring of the founder of OpenClaw, Microsoft sales leadership emailing staff to pitch against OpenAI's new agent platform, Salesforce privately retreating from pure LLM reasoning in Agentforce, and a dozen startups founded by former OpenAI and DeepMind researchers raising at billion-dollar valuations to build agents for narrow verticals. I've been covering tech for more than two decades. I have not seen a gap between the public narrative and the business reality this wide since the early days of crypto. The public narrative says general-purpose AI agents that do everything are coming fast. The reporting tells a different story. Enterprises are losing trust in LLMs, not gaining it. The biggest platform companies are already fighting over who controls the layer between the model and the customer. (That's what those Microsoft internal emails were really about.) And the smartest money isn't going into do-everything agents. It's going into narrow ones trained on licensed data from specific industries — companies like Applied Compute (raising at a $1.3 billion valuation for legal AI) and Elorian (visual reasoning agents from ex-DeepMind researchers). That last part is the one I keep coming back to. OpenAI and Anthropic are actively courting biotech and financial firms to license proprietary data — genomics, tax records, code bases — to train agents that can do expert-level work in those fields. The companies making these deals right now are shaping competitive dynamics that will play out for years. The agent story isn't really a product story. It's a story about how specialized knowledge gets restructured across entire industries. And the details—who is selling what data, to whom, at what price, under what terms—matter enormously. That is the kind of reporting we built The Information to do. If you have ever considered subscribing to The Information, now is the time. https://lnkd.in/gNZJNJpM

I passed on 473 AI startups last year who claimed proprietary models or data. None had actual moats. As someone investing in AI with data moats, I can spot the pretenders instantly. What founders claim: "We have a proprietary AI model" - Built on OpenAI API? "We have unique training data" - 1,000 data points they scraped? And the data points aren't directly relevant to their agentic use-cases. "We have technical advantages" - Using same tech as everyone else? Your RAG isn't a technical advantage. It's just your application's business logic. "We fine-tuned for our vertical" - and then in diligence I discover it’s just better RAG... fine-tuning requires really data science work. None of these are moats. The brutal truth about AI startups: 95% are thin wrappers on OpenAI/Anthropic APIs. When GPT-6 or Claude 5 comes out, your "advantage" disappears. When OpenAI launches native feature, you're dead. Your "proprietary" is replicable in 3 months. One founder: "We built a proprietary model for legal research." Me: "What's it built on?" Founder: "We fine-tuned Llama on legal documents." Me: "How many documents?" Founder: "About 500." Me: "What happens when OpenAI releases GPT-6 with better legal reasoning on 100k documents?" Founder: "Well... uh..." No moat. Another: "We have proprietary data for AI training." Me: "How much data? Is the data applicable to the use-cases you are solving for?" Founder: "We've collected 50,000 data points over 6 months." Me: "How long would it take a competitor to replicate? Is it available to everybody?" Founder: "Probably... 6 months? Yes, you can scrap the data online" That's not proprietary. That's a head start. Real AI moats are rare: Moat 1: Proprietary data that can't be replicated Years of data nobody else has today Legally protected or exclusive via partnership Moat 2: Network effects in data Each incremental user improves model for all users Each incremental agent launched and course correct by users builds more reinforcement learning data improving agents Competitor starting from zero has inferior product Moat 3: Vertical integration that's prohibitively expensive Own the full stack from data collection to deployment Data derived from proprietary Hardware Requires $50M+ to replicate. A great example is autonomous vehicle companies with proprietary sensor data + fleet Moat 4: Regulatory capture Exclusive rights through regulation Years to replicate approvals If you're building AI startup: Stop claiming your model is proprietary unless you truly can't be replicated for years. Focus on distribution. Build actual moat: Data, network effects, or regulatory. Or accept you're a feature, not a company, and build accordingly. The AI gold rush is creating hundreds of companies that won't exist in 3 years. Don't be one of them. #AI #StartupStrategy #VentureCapital

With the merger of xAI and X, Elon Musk has done something critical in today’s AI race: he’s secured a data moat. In an era where Ilya Sutskever warns that “we have but one internet,” access to proprietary, high-quality, real-world data is becoming the true differentiator—not just model size or architecture. While OpenAI and Anthropic often lead the narrative on innovation and model capabilities, ironically, their data moats are easier to cross. They rely heavily on publicly available data and curated datasets, which are nearing exhaustion. Meanwhile, Meta (with Facebook and Instagram), Google (with search, YouTube, and Gmail), and now Musk (with X) are locking down massive, natural-language data streams at scale. These aren’t just competitive advantages—they’re defensible ecosystems. In the AI arms race, the real question isn’t “Who has the best model?” It’s: Who owns the best data?

🚨 Most “AI” isn’t AI. There, I said it. No legal definition + no regulatory precedent = free market. That means vendors can slap "AI-powered" or "Proprietary AI" on anything. In reality? Most of what’s being sold is not proprietary And is just automation flows with a public LLM sprinkled on top. So, if you want to expose the truth (and protect yourself from legal and compliance risks), ask these three questions: 🔍 1. What data was your model trained on? The answer should be first-party data or legally acquired proprietary datasets. If they mention “public data,” that’s not a free pass. Push further. Did the individuals whose data was scraped know? If not, there’s a compliance risk in the UK, EU, and US. 🔍 2. How do you ensure explainability and mitigate bias? A real AI model should have a structured process for detecting and reducing bias. That means documented audits, third-party testing, or compliance with AI risk frameworks like the EU AI Act. If their answer is “we continuously improve our model,” that’s not a process—it’s an excuse. Bias lawsuits and regulatory scrutiny are already happening. If they can’t clearly explain how they control risk, they pass that risk onto you. 🔍 3. Can your "proprietary AI" operate without third-party APIs? If they hesitate or say no, they’re not selling proprietary AI They’re selling automation wrapped around OpenAI, Anthropic, or another LLM. That means you’re exposed to the same privacy and compliance risks as using those models directly. Ask what contractual protections they have. Are they indemnifying you if the LLM provider makes changes, goes down, or faces legal action? If they can’t answer, you’re not just buying software—you’re buying uncertainty. 💡 Why this matters: Regulations are tightening and compliance isn’t optional. If a vendor can’t answer these questions clearly, you could be buying a lawsuit waiting to happen. Ask the questions. Document the answers. And if you catch someone squirming… You'll know to move on to the next option. #AI #Ethics #DataLaw 📸: Scooby-Doo villain reveal meme—except when the mask comes off, AI is just a simple set of rules underneath.

Is 𝗩𝗶𝗯𝗲 𝗖𝗼𝗱𝗶𝗻𝗴 𝗘𝘅𝗽𝗼𝘀𝗶𝗻𝗴 𝗬𝗼𝘂𝗿 𝗖𝗼𝗺𝗽𝗲𝘁𝗶𝘁𝗶𝘃𝗲 𝗘𝗱𝗴𝗲 and IP? (Post 6 of ~27 in my Public AI Risk Series) Did you know that 𝗽𝗮𝘀𝘁𝗶𝗻𝗴 𝗽𝗿𝗼𝗽𝗿𝗶𝗲𝘁𝗮𝗿𝘆 𝘀𝗼𝘂𝗿𝗰𝗲 𝗰𝗼𝗱𝗲, 𝘂𝗻𝗶𝗾𝘂𝗲 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺𝘀, 𝗼𝗿 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝗜𝗣 𝗶𝗻𝘁𝗼 𝗮 𝗽𝘂𝗯𝗹𝗶𝗰 𝗟𝗟𝗠 is a big risk. Bug fixing is, too. It sounds harmless: "Can you help me debug this?" But here's reality: 𝗬𝗼𝘂'𝗿𝗲 𝘁𝗿𝘂𝘀𝘁𝗶𝗻𝗴 𝘃𝗲𝗻𝗱𝗼𝗿 𝗽𝗼𝗹𝗶𝗰𝗶𝗲𝘀 𝘁𝗼 𝗽𝗿𝗼𝘁𝗲𝗰𝘁 𝘆𝗼𝘂𝗿 𝗰𝗼𝗺𝗽𝗲𝘁𝗶𝘁𝗶𝘃𝗲 𝗲𝗱𝗴𝗲 — 𝗮𝗻𝗱 𝘁𝗵𝗼𝘀𝗲 𝗽𝗼𝗹𝗶𝗰𝗶𝗲𝘀 𝗰𝗮𝗻 𝗰𝗵𝗮𝗻𝗴𝗲 𝗼𝘃𝗲𝗿𝗻𝗶𝗴𝗵𝘁. Once your code hits their logs, you no longer control where it lives or how it's used. Why it's a 𝗿𝗶𝘀𝗸: Public AI creates multiple exposure points: ⛔ Retention. Your code resides in vendor logs you don't control. ⛔ Training potential. Vendors can — and do — change training policies. ⛔ Model inversion. Rare, but proven: structural logic can be reconstructed through careful querying. And there's more (see the comments). Put simply: 𝗜𝗳 𝘆𝗼𝘂𝗿 𝗮𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 𝗼𝗿 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺𝘀 𝘁𝗼𝘂𝗰𝗵 𝗮 𝗽𝘂𝗯𝗹𝗶𝗰 𝗺𝗼𝗱𝗲𝗹, 𝘁𝗵𝗲𝘆 𝗺𝗮𝘆 𝗹𝗲𝗮𝘃𝗲 𝗳𝗶𝗻𝗴𝗲𝗿𝗽𝗿𝗶𝗻𝘁𝘀 — 𝗶𝗻 𝗹𝗼𝗴𝘀, 𝗽𝗮𝘁𝘁𝗲𝗿𝗻𝘀, or 𝗳𝘂𝘁𝘂𝗿𝗲 𝗺𝗼𝗱𝗲𝗹 𝗯𝗲𝗵𝗮𝘃𝗶𝗼𝗿. This is how IP can leak without anyone noticing. Think of it like this: Using public AI to debug company code is like bringing a prototype engine into a crowded workshop and asking strangers for advice. They'll help… but now everyone knows how your engine works, what makes it special, and what makes it fast. And some of those people might be building engines, too. 𝗔 𝗯𝗶𝗴 𝗺𝗶𝘀𝗰𝗼𝗻𝗰𝗲𝗽𝘁𝗶𝗼𝗻: People think: "I deleted the chat, so it's gone." Not even close. The deletion button clears your view — not the vendor's logs, not the retention systems, not the audit trails. It's a bit like deleting messages off your cell phone. 𝗧𝗵𝗲 𝗰𝗼𝗿𝗲 𝗶𝘀𝘀𝘂𝗲: 𝗣𝘂𝗯𝗹𝗶𝗰 𝗔𝗜 𝗮𝗻𝗱 𝗽𝗿𝗼𝗽𝗿𝗶𝗲𝘁𝗮𝗿𝘆 𝗰𝗼𝗱𝗲 𝗱𝗼 𝗻𝗼𝘁 𝗺𝗶𝘅. It's your secret sauce. Your differentiation. Your advantage. Why risk that on infrastructure where retention, access, and training decisions sit outside your control? 𝗧𝗵𝗲 𝘀𝗼𝗹𝘂𝘁𝗶𝗼𝗻? Use AI. But use Private AI. Private AI for proprietary work. Private AI lives in your environment, under your security, following your rules. No vendor logs. No policy surprises. Just sovereignty over your IP. 𝗧𝗮𝗸𝗲𝗮𝘄𝗮𝘆 𝗡𝗲𝘃𝗲𝗿 𝗽𝗮𝘀𝘁𝗲 𝘀𝗼𝘂𝗿𝗰𝗲 𝗰𝗼𝗱𝗲 𝗼𝗿 𝘂𝗻𝗶𝗾𝘂𝗲 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺𝘀 𝗶𝗻𝘁𝗼 𝗽𝘂𝗯𝗹𝗶𝗰 𝗔𝗜. It's not just code — it's your competitive edge. 🌐 𝗣𝘂𝗯𝗹𝗶𝗰 𝗔𝗜 𝗰𝗿𝗲𝗮𝘁𝗲𝘀 𝗿𝗶𝘀𝗸. 𝗣𝗿𝗶𝘃𝗮𝘁𝗲 𝗔𝗜 𝗽𝗿𝗼𝘁𝗲𝗰𝘁𝘀 𝗶𝘁. I've added a simple explainer in the comments. #AI #Cybersecurity #PrivateAI #DataSecurity #DevSecOps

Here's a question: Why are so many businesses using the exact same off-the-shelf AI tools as their direct competitors and expecting to gain a unique advantage? A real, sustainable competitive edge doesn't come from a shared product. It comes from building your own intellectual property. This is the fundamental difference between 'renting' a generic AI and owning a bespoke one. When you build a custom AI, it’s trained on your most valuable asset: your proprietary data. Your internal process logs, your unique customer interaction history, your specific performance metrics. This is a goldmine that generic tools simply cannot access or understand. Let’s make this practical. Imagine a UK manufacturing firm struggling with machinery downtime. They try a generic predictive maintenance tool. It fails. Why? Because it can't integrate with their proprietary sensors or understand the unique operational stresses of their specific machinery. With a bespoke solution, you build an AI that: ✅ Integrates perfectly with their existing legacy SCADA systems. ✅ Is trained exclusively on their years of historical performance data (vibration patterns, temperature, etc.). ✅ Understands the specific failure signatures of their machines. The result isn't a generic dashboard. It's a pinpoint-accurate prediction that a critical component will fail in three days. Maintenance is scheduled, production isn't disrupted, and the business saves a fortune. That is an advantage your competitors cannot copy. That’s your secret weapon. Read more on our new blog: https://lnkd.in/eHk4tD42 If you could build an AI to solve just one unique, high-value problem in your business, what would it be? #BespokeSoftware #PredictiveMaintenance #AIforManufacturing

That AI chat? A court just ruled it's not confidential. A federal judge just ruled that 31 documents a defendant created using Claude AI aren't protected by attorney-client privilege. The reasoning was straightforward: Claude isn't your lawyer. Anthropic owes you no duty of confidentiality. Their terms permit government access. The moment you typed your strategy into that chat window, you disclosed it to a third party. Privilege gone. And it doesn't matter if you're on the free plan or paying $20/month. Consumer terms are identical across tiers. Sam Altman said it plainly: "There's no legal confidentiality when using ChatGPT." Same applies to Claude. Same applies to Gemini. Enterprise versions of these tools do offer stronger protections, no-training clauses, data processing agreements, explicit confidentiality provisions. But most employees aren't using the enterprise version. They're using the same consumer tools the rest of us are. But here's what most people will miss. This isn't just a legal profession problem. How many of your employees are uploading financial models, client lists, and proprietary data into consumer AI tools right now? They're trying to work faster. But if a court says inputting sensitive information into a commercial AI equals disclosure to a third party, and one just did, every one of those uploads is a potential breach of an NDA, employment agreement, or trade secret protection. And once it's in? You can't get it back. Stanford researchers have documented that removing data from a trained model is "nearly impossible." Metadata, prompt logs, agent traces, they persist even after you hit delete. This is a red light for every CEO and board member. Not just the lawyers. The question isn't whether your people are using AI. They are. The question is whether you've told them what's safe to put in and what isn't.

🧨 The real AI arms race isn’t about models. It’s about data—and if you're not paying attention, you're already losing. For the last few years, AI has been all about whose LLM is bigger, faster, or more powerful. But models are rapidly commoditizing. Open-source alternatives (Mistral, LLaMA, Mixtral) are catching up fast. And cloud providers have made swapping AI models as easy as changing a line of code. Companies can now: Run micro-models locally for real-time applications Deploy in-house models for security & cost reasons Call cloud-hosted models (OpenAI, Anthropic, open-source LLMs) The model isn’t the moat anymore. The data is. Who’s already winning on data? The most valuable AI companies aren’t just great at training models—they own unique proprietary datasets that no one else can access. 🔹 Waymo’s self-driving data → 20M+ real-world miles, 20B+ simulated miles. Anyone can build a self-driving model. No one else has this data. 🔹 Google’s search intent data → 20+ years of user behavior, fine-tuned to what people actually want. 🔹 Epic’s Fortnite metaverse data → AI-generated characters & environments are improving fast—but Epic has millions of hours of real 3D interaction data. Where the next AI data moats will be built If models are commoditizing, the next AI giants will own new, untapped proprietary datasets. These are some of the highest-value areas still up for grabs: 🔹 AI-generated code execution data → The key to self-improving AI coding copilots. 🔹 Real-world human decision-making → AI that actually understands why experts make choices. 🔹 Full-body motion tracking → The missing dataset for AI-powered robotics & industrial automation. 🔹 AI-generated molecule interactions → Unlocking new drugs, materials, and biotech breakthroughs. 🔹 Emotional & persuasive language mapping → AI that adapts its communication style to different people. What this means for founders Most AI startups are competing on models. The smartest ones are securing data moats. The real question isn’t “What’s your model?”—it’s “What’s your unique dataset?” If you were starting an AI company today, what’s the one dataset you’d want to own? Drop your thoughts below. ✍️

At the end of the day, it all comes back to DATA. If everyone is using the same foundation models, the only real competitive edge left is your own enterprise data. Your proprietary data is what turns a general-purpose model into YOUR model. A model that reflects your customers, your industry, your unique challenges and opportunities, your proprietary knowledge and experience. The organizations that win today won’t be those who plug into the latest LLM. They’ll be the ones who: (1) curate high-quality, domain-specific data, (2) invest in data infrastructure and governance and (3) continually fine-tune and adapt models with their own insights. The fundamentals haven’t changed: the value is in the data. Those who recognize this and treat data as their most strategic asset will be the ones who lead in the age of AI. #data #artificialintelligence #generativeAI #strategy #LLMs TL;DR Don't chase GenAI hype without addressing your core data reality first. Trusted data (aka cleaned, governed, AI-ready data) is your real differentiator.

84% say data is their AI edge. Only 26% trust it. From 1,700 CDOs across 27 countries. That gap is where AI ROI collapses. And where most enterprises fall behind. The IBM CDO Study makes one thing clear. Companies don’t fail because their models are weak. They fail because their data is unusable. Here is what actually matters 👇 1️⃣ 𝗗𝗔𝗧𝗔 𝗔𝗦 𝗣𝗥𝗢𝗗𝗨𝗖𝗧 Proprietary data creates advantage only when it is packaged. Reusable. Interoperable. Agent ready. Customer signals. Operational telemetry. Financial and risk signals. Unstructured conversations and documents. Leaders turn these into data products with clear ownership and distribution. This is how proprietary data becomes defensible. 2️⃣ 𝗣𝗜𝗣𝗘𝗟𝗜𝗡𝗘𝗦 𝗙𝗢𝗥 𝗔𝗜 𝗔𝗚𝗘𝗡𝗧𝗦 The study shows the core barrier to AI scale is fragmentation. Silos. Inconsistent taxonomies. Incomplete lineage. Slow access. AI multiplies value only when data is engineered for flow. Decision ready. Context rich. Available in real time to both humans and agents. Pipelines replace warehouses. Movement replaces storage. Flow replaces accumulation. 3️⃣ 𝗚𝗢𝗩𝗘𝗥𝗡𝗔𝗡𝗖𝗘 𝗔𝗦 𝗔𝗖𝗖𝗘𝗦𝗦 Leaders converge on one idea. The risk of restricting data now exceeds the risk of sharing it with controls. Federated access. Role based models. AI agent marketplaces. Guardrails without bottlenecks. This is how organizations convert proprietary data into competitive motion. 𝗠𝗬 𝗧𝗔𝗞𝗘𝗔𝗪𝗔𝗬 The next winners will not be the companies with the most data. They will be the companies that turn their proprietary data into 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝘀 𝗮𝗴𝗲𝗻𝘁𝘀 𝗰𝗮𝗻 𝘂𝘀𝗲 𝗶𝗺𝗺𝗲𝗱𝗶𝗮𝘁𝗲𝗹𝘆. This is the real competitive moat. And most enterprises are years behind. The AI multiplier appears only when data stops being stored and starts being operationalized. 👉 If you’re defining your AI strategy, let’s talk. I help leaders turn data into operating models that scale intelligent workflows across the enterprise. Save 💾 React 👍 Share ♻️ Follow