# Datasaur

> Datasaur is a private AI platform for regulated industries. We build and run secure, model-agnostic AI inside your own infrastructure (VPC, on-premise, or air-gapped), so your data stays in your environment and is never used to train third-party models. Not a data-labeling tool: a private AI platform. Data Studio, our data-labeling product, is one part of the pipeline behind those deployments.

Datasaur serves healthcare, legal, finance, insurance, and government teams that cannot send data to someone else's cloud. Two offerings: Forge, an AI-native services practice that deploys private AI into your environment, and the Datasaur platform (LLM Labs and Data Studio) for building, evaluating, and feeding those models.

## Core
- [Private LLMs & Secure Enterprise AI](https://datasaur.ai/): What Datasaur does: private AI deployed inside your own infrastructure.
- [FAQ](https://datasaur.ai/faq): Common questions about private AI deployment, security, and models.
- [Data Privacy](https://datasaur.ai/data-privacy): How Datasaur handles data privacy, residency, and retention.
- [Enterprise](https://datasaur.ai/enterprise): Enterprise deployment, security, and support model.
- [Datasaur Forge](https://datasaur.ai/forge): Forge, the AI-native services practice that builds and runs private AI in your environment.
- [Pricing | Forge](https://datasaur.ai/forge/pricing-forge): Forge engagement pricing.
- [Explore Product | Forge](https://datasaur.ai/forge/explore-product-forge): What a Forge engagement delivers.
- [Private AI Workflows & Approach](https://datasaur.ai/product): How private AI deployment works, from model selection to production.
- [Enterprise Pricing](https://datasaur.ai/pricing): Enterprise pricing tiers, from a single workflow to a strategic partnership.

## Private AI and Sovereign AI
- [Custom LLM Platform](https://datasaur.ai/custom-llm): Custom LLM platform for building and tuning models on your own data.
- [Build Your Own LLM](https://datasaur.ai/llm/build-your-own-llm): Build your own LLM: model selection, tuning, evaluation, and deployment.
- [Private LLMs for Regulated Enterprises](https://datasaur.ai/llm/private-llm-2): Private LLMs for regulated enterprises: what they are and how Datasaur deploys them.
- [Custom LLM](https://datasaur.ai/llm/custom-llm): Build a custom LLM tuned on your own data, deployed in your environment.
- [Local LLM](https://datasaur.ai/llm/local-llm): Run open-weight LLMs entirely on your own hardware, with no data leaving your environment.
- [Custom GPT](https://datasaur.ai/llm/custom-gpt): Private, custom GPT-style assistants that run inside your infrastructure.
- [Cohere Alternative](https://datasaur.ai/llm/cohere-alternative): Cohere alternative: private, model-agnostic AI you host yourself.
- [Accenture Alternative](https://datasaur.ai/llm/accenture-alternative): Accenture alternative: an AI-native services practice, not a generalist consultancy.
- [Harvey AI Alternative](https://datasaur.ai/llm/harvey-ai-alternative): Harvey AI alternative: private legal AI deployed in your own environment, model-agnostic.
- [Sovereign AI](https://datasaur.ai/llm/sovereign-ai-datasaur): Sovereign AI: LLMs deployed inside your own infrastructure, on-premise or air-gapped, aligned to your data residency and sovereignty policy.

## Industries
- [Responsible AI for Legal Services](https://datasaur.ai/industries/legal): Private AI for legal teams: matter analysis and document review without sending client data to public LLMs.
- [HIPAA-Compliant AI for Healthcare](https://datasaur.ai/industries/healthcare): HIPAA-compliant private AI for healthcare: claims, intake, and record analysis without exposing PHI to public models.
- [Compliant AI for Financial Services](https://datasaur.ai/industries/finance): Private AI for financial services: defensible intelligence from financial data, inside your controls.
- [Compliant Private AI for Insurance Carriers](https://datasaur.ai/industries/insurance): Private AI for insurance carriers: risk selection and claims adjudication on secure infrastructure.
- [AI That Scales With Your E-Commerce Operations](https://datasaur.ai/industries/ecommerce): Private AI for eCommerce operations, keeping first-party data in your environment.
- [Secure AI Built for Government Agencies](https://datasaur.ai/industries/government): Sovereign AI for government agencies: air-gap support, full auditability, data never leaves your infrastructure.

## Thinking
- [Blog](https://datasaur.ai/blog): Index of all posts. Posts on fast-moving AI/model topics may be superseded by newer ones; check publish dates.
- [AI Compute Cost Runaway: Why Your Token Bill Is Only Going Up](https://datasaur.ai/blog/ai-compute-cost-runaway-why-your-token-bill-is-only-going-up): Datasaur's AI coding spend is up over 400% in 18 months, and the cost curve is unlike anything in enterprise software because the better the tools get, the more teams use them with no natural...
- [AI Is Coming for Mathematics, and the Rest of Science Is Next](https://datasaur.ai/blog/ai-is-coming-for-mathematics-and-the-rest-of-science-is-next): OpenAI disproved an 80-year-old geometry conjecture and Google DeepMind solved 9 open Erdős problems in a single weekend, each reportedly costing only hundreds of dollars to compute.
- [Building the Backend for AI Agents with Datasaur](https://datasaur.ai/blog/building-the-backend-for-ai-agents-with-datasaur): Datasaur empowers AI Agent development with tools for model evaluation, data labeling, and secure deployment, ensuring optimized performance and continuous improvement.
- [Chinese Open-Weights Models: Security Myths vs. Reality](https://datasaur.ai/blog/chinese-open-weights-models-security-myths-vs-reality): Organizations worldwide are increasingly leveraging powerful open-weights AI models, such as Alibaba’s Qwen and DeepSeek’s R-series, to drive innovation.
- [Court orders OpenAI to Retain Data](https://datasaur.ai/blog/court-orders-openai-to-retain-data): A recent court order now requires OpenAI to retain and segregate most ChatGPT user conversations, raising fresh concerns about data privacy and security.
- [Datasaur Launches Forge, an AI Native Service for Private, Enterprise AI](https://datasaur.ai/blog/datasaur-launches-forge-an-ai-native-service-for-private-enterprise-ai): Datasaur is launching Forge, an AI Native Service that embeds engineers directly inside regulated organizations to build and operate AI systems that run entirely within the customer's own...
- [Enterprise-Grade AI Without Compromise: How Datasaur Delivers a Secure, Private LLM](https://datasaur.ai/blog/enterprise-grade-ai-without-compromise-how-datasaur-delivers-a-secure-private-llm): The use of of LLMs is exploding. Many enterprise companies have hesitated to adopt the technology due to security and privacy reasons. Datasaur resolves these concerns in multiple ways.
- [From “Agents” to Autonomy: A Practical Framework for Agentic AI (Levels 1–5)](https://datasaur.ai/blog/from-agents-to-autonomy-a-practical-framework-for-agentic-ai-levels-1-5): Agentic AI isn’t just a binary “bot or not”, it lives on a spectrum of autonomy. There are five levels: from basic rule-based helpers, to tool-using assistants, up to fully autonomous agents that can...
- [GitHub's 30X Wake-Up Call: The Agent Compute Tsunami Is Coming](https://datasaur.ai/blog/githubs-30x-wake-up-call-the-agent-compute-tsunami-is-coming): GitHub just revised its capacity needs from 10X to 30X in just four months and that's only early adopters. The agentic AI wave hasn't even hit mainstream yet.
- [Google I/O 2026: The Enterprise CIO's Cheat Sheet](https://datasaur.ai/blog/google-i-o-2026-the-enterprise-cios-cheat-sheet): What enterprise CIOs need to know from Google I/O 2026, distilled into a practical cheat sheet.
- [Governing Claude Code in the Enterprise](https://datasaur.ai/blog/governing-claude-code-in-the-enterprise): Agentic tools like Claude Code do not just generate text, they can execute commands, access files, and interact with production systems, and most enterprise governance frameworks were not designed...
- [I Built My Own Morning Podcast with Three Free Tools](https://datasaur.ai/blog/i-built-my-own-morning-podcast-with-three-free-tools): A personalized AI morning briefing that reads you your Slack messages and emails aloud, delivered to Spotify like a daily podcast, built in a weekend using three free tools.
- [Match the Model to the Job](https://datasaur.ai/blog/match-the-model-to-the-job): Most enterprises do not have an AI problem, they have an allocation problem, routing every workflow through the most expensive model regardless of whether the task justifies the cost.
- [Open Models Are Winning And Enterprises Are About to Find Out](https://datasaur.ai/blog/open-models-are-winning-and-enterprises-are-about-to-find-out): Open-source models now account for the majority of token traffic on OpenRouter, not because users lack access to frontier APIs, but because informed practitioners with access to both are actively...
- [Open Weights Models Are Having Their Moment: A Field Report](https://datasaur.ai/blog/open-weights-models-are-having-their-moment-a-field-report): Three open-weight model releases in a single week, from the US, China, and Google, now cover every major axis of enterprise AI decision-making: raw performance, cost at scale, and local deployment.
- [Private LLMs: Definition, Spectrum, and a Buyer’s Framework](https://datasaur.ai/blog/private-llms-definition-spectrum-and-a-buyers-framework): A Private LLM enforces your organization’s privacy and compliance end-to-end while being customized to your workflows and optimized for quality, latency, and cost—with clear auditability and...
- [The AI Ownership Threshold](https://datasaur.ai/blog/the-ai-ownership-threshold): The gap between the flagship you rent and the model you can own has compressed to weeks, with Opus 5, Sol and Kimi K3 now within a point of each other on agentic coding.
- [The Best Model You Can Actually Deploy Is the One You Control](https://datasaur.ai/blog/the-best-model-you-can-actually-deploy-is-the-one-you-control): The most powerful model on the market is not always the most deployable one, and for regulated enterprises, a 30-day prompt retention policy is not a minor detail, it is a hard stop.
- [The Deployment Layer Is Where Enterprise AI Is Won](https://datasaur.ai/blog/the-deployment-layer-is-where-enterprise-ai-is-won): Anthropic and OpenAI announced billion-dollar services arms within twelve hours of each other, and the signal is clear: frontier model capability is no longer the competitive edge in enterprise AI.
- [The Model Is Replaceable. The Learning Loop Is the Moat](https://datasaur.ai/blog/the-model-is-replaceable-the-learning-loop-is-the-moat): Satya Nadella made it explicit: the frontier model is a swappable input, not a moat. The enterprises that will compound their AI advantage are the ones building private evaluation loops and learning...
- [The Private AI Stack Has Arrived: What Dell and NVIDIA's AI Factory 2.0 Means for Enterprise Buyers](https://datasaur.ai/blog/the-private-ai-stack-has-arrived-what-dell-and-nvidias-ai-factory-2-0-means-for-enterprise-buyers): Dell and NVIDIA announced turnkey on-premise AI infrastructure with an 87% cost reduction versus public cloud APIs. What it means for enterprise buyers evaluating private AI.
- [The Token ROI Attribution Gap: Why Spending Caps Are the Wrong Answer](https://datasaur.ai/blog/the-token-roi-attribution-gap-why-spending-caps-are-the-wrong-answer): Most organizations know that tokenmaxxing is wasteful, but almost none can define what good token consumption actually looks like, and that gap is where AI cost governance goes wrong.
- [Want to Use ChatGPT at Work But Can't?](https://datasaur.ai/blog/want-to-use-chatgpt-at-work-but-cant): Explore the various ways your company's data can be exposed through the use of LLMs and, more importantly, what you can do to protect it.
- [What the Next AI Breakthrough Looks Like: Scale, Context, and Open Source](https://datasaur.ai/blog/what-the-next-ai-breakthrough-looks-like-scale-context-and-open-source): A likely next “DeepSeek moment” in AI could come from the convergence of massive scale (1.6T parameters), radically larger context windows (up to 1M tokens), and permissive open-source licensing...
- [When Surge Pricing Came for AI: Why Private LLMs Are the Enterprise Answer](https://datasaur.ai/blog/when-surge-pricing-came-for-ai-why-private-llms-are-the-enterprise-answer): Private LLMs let enterprises run AI inside their own perimeter, keeping sensitive data, predictable costs, and uptime under their control.
- [When You Rent the Frontier, You Control Nothing That Matters](https://datasaur.ai/blog/when-you-rent-the-frontier-you-control-nothing-that-matters): Anthropic's Fable launch lasted three days before access changed, data retention terms shifted without warning, and users discovered some prompts were silently rerouted to a different model entirely.
- [When Your 2026 Budget Is Gone by April](https://datasaur.ai/blog/when-your-2026-budget-is-gone-by-april): Uber's CTO reportedly burned through the entire 2026 tech budget in just four months, a sign that traditional annual planning can't keep pace with how fast modern tech costs evolve.
- [Why Enterprises Are Building Their Own AI And Why It's Just Getting Started](https://datasaur.ai/blog/why-enterprises-are-building-their-own-ai-and-why-its-just-getting-started): Kirkland & Ellis is committing $500 million to build proprietary AI because when every firm in your industry uses the same tools, those tools stop being an advantage and become the floor.
- [Why I’m More Excited About Dreaming Than 2x Token Limits](https://datasaur.ai/blog/why-im-more-excited-about-dreaming-than-2x-token-limits): Anthropic's doubled token limits are getting all the attention, but the more consequential announcement is Dreaming, a capability that lets agents review their own sessions and rewrite their own...
- [Why Local AI Is Winning: The Localmaxxing Benchmark That Changes the Conversation](https://datasaur.ai/blog/why-local-ai-is-winning-the-localmaxxing-benchmark-that-changes-the-conversation): A new benchmark shows a local model on a laptop outrunning Anthropic's Opus 4.5 by 2x on latency for routine agentic tasks, with no meaningful accuracy difference.
- [Why Owning Your AI Deployment Stack Beats Renting It](https://datasaur.ai/blog/why-owning-your-ai-deployment-stack-beats-renting-it): OpenAI's $10 billion DeployCo bet confirms deployment, not model access, is the hard part of enterprise AI. Why owning your stack beats renting it.
- [Why Regulated Industries Will Start Owning Their Frontier AI Models](https://datasaur.ai/blog/why-regulated-industries-will-start-owning-their-frontier-ai-models): Mayo Clinic is building a frontier AI model trained on its own clinical data that it will own outright, and that is a fundamentally different move than any enterprise API arrangement.
- [You Can Now Run Any LLM Inside Claude Code](https://datasaur.ai/blog/you-can-now-run-any-llm-inside-claude-code): Anthropic quietly added the ability to load any model into Claude Code, including Qwen, GPT-5.5, and Grok, with no announcement or blog post.
- [You Don’t Own the Off Switch](https://datasaur.ai/blog/you-dont-own-the-off-switch): When a government can cut access to a frontier model overnight, it exposes something most teams treat as an implementation detail: the API you depend on is not your infrastructure, it belongs to...
- [Your SaaS Stack Is Leaking Data to AI Models You Never Approved](https://datasaur.ai/blog/your-saas-stack-is-leaking-data-to-ai-models-you-never-approved): Nearly two thirds of SaaS vendors are actively sending your company's data to AI models you never approved, and the AI addendum your legal team negotiated only binds the vendor who signed it, not the...

## Data Studio and LLM Labs
- [About Datasaur | Leading NLP Labeling for AI and ML](https://datasaur.ai/studio/about-general): About Data Studio: the team and story behind Datasaur's labeling platform.
- [e-Commerce Industry NLP Labeling Solutions](https://datasaur.ai/studio/ecommerce): AI-driven chatbots, sentiment analysis, customer service reviews, and more. Datasaur is the best text annotation tool to get you there.
- [Financial Industry NLP Labeling Solutions](https://datasaur.ai/studio/financial): Datasaur is the most robust, customizable NLP labeling tool for Finance and Fintech. Automate 80% of labeling and reduce project times by 10X. Find out more.
- [Healthcare Industry NLP Labeling Solutions](https://datasaur.ai/studio/healthcare): Customizable, advanced NLP data labeling to drive innovation for healthcare organizations. Learn more and set up a custom demo.
- [Legal NLP Solutions](https://datasaur.ai/studio/legal): NLP labeling for legal text: contracts, case documents, and matter data, in Data Studio.
- [Media Industry NLP Labeling Solutions](https://datasaur.ai/studio/media): NLP labeling for media and publishing content: transcripts, articles, and editorial workflows, in Data Studio.
- [Interactive Playground](https://datasaur.ai/studio/playground-general): Try Data Studio's labeling tools on sample text in an interactive playground.
- [Pricing - Data Studio](https://datasaur.ai/studio/pricing): Data Studio pricing.
- [Compare NLP Labeling Solutions](https://datasaur.ai/studio/compare): How Data Studio compares to other NLP annotation tools.
- [Complete NLP Labeling Solutions](https://datasaur.ai/studio/explore-nlp-product): Easily label conversations, essays, or any text-based documents rapidly with customizable automation using Datasaur.
- [Automations for NLP Labeling](https://datasaur.ai/studio/automations): Datasaur offers numerous ways to automate your data-labeling process, saving our users up to 80% of their time and effort.
- [Data Studio](https://datasaur.ai/studio): Data Studio, Datasaur's NLP data-labeling platform. Part of the data pipeline behind our private AI deployments.
- [LLM Labs](https://datasaur.ai/studio/llm/home): LLM Labs: build, compare, and evaluate models before deploying them privately.
- [Craft And Evaluate Your LLM](https://datasaur.ai/studio/llm/explore-product): Craft and evaluate LLMs in LLM Labs.
- [Healthcare Industry Solutions Powered by LLM Labs](https://datasaur.ai/studio/llm/healthcare): LLM Labs for clinical documentation and intake: compare models on your data before a private deployment touches PHI.
- [Legal | LLM](https://datasaur.ai/studio/llm/legal): Evaluate LLMs on contract review and case research in LLM Labs, then deploy the winner privately inside your firm's environment.
- [Government NLP Labeling Solutions](https://datasaur.ai/studio/government): NLP labeling for government and public-sector text data, with compliance-aware workflows, in Data Studio.
- [Government Solutions Powered by LLM Labs](https://datasaur.ai/studio/llm/government): Model evaluation for public-sector casework and records, run in LLM Labs ahead of an air-gapped or on-premise rollout.
- [Pricing - LLM Labs](https://datasaur.ai/studio/llm/pricing): LLM Labs pricing.
- [Pricing](https://datasaur.ai/studio/pricing-general): Datasaur pricing overview. See /pricing for current enterprise tiers.
- [AI Development Services](https://datasaur.ai/studio/llm/ai-development-services): AI development services delivered on LLM Labs.
- [Private LLM (LLM Labs)](https://datasaur.ai/studio/llm/private-llm): Private LLM deployment through LLM Labs.
- [Create Your Own AI](https://datasaur.ai/studio/llm/create-your-own-ai): Create your own AI assistant on your own data.
- [LLM App](https://datasaur.ai/studio/llm/llm-app): Build LLM-backed applications.
- [LLM Tool](https://datasaur.ai/studio/llm/llm-tool): LLM tooling for evaluation and comparison.
- [Playground | NLP Labeling](https://datasaur.ai/studio/playground): Datasaur supports many types of text labeling. Try out the Datasaur platform in our Playground and see how easy and intuitive it is for yourself!
- [Guides | NLP Labeling](https://datasaur.ai/studio/guides): Building quality NLP models is complex, so we’ve created these guides to deliver maximum business impact and streamline team efficiency. Find the guides here.
- [Integrations | NLP Labeling](https://datasaur.ai/studio/integrations): Data Studio integrations: connect labeling workflows to your existing ML stack.
- [About | NLP Labeling](https://datasaur.ai/studio/about): Datasaur is backed by leading investors and VCs to bring the best text and audio annotation tool to the market.
- [Blog | NLP Labeling](https://datasaur.ai/studio/blog): Gain industry insights, product updates, Datasaur platform comparisons, and more with our data labeling blog.
- [Case Studies | NLP Labeling](https://datasaur.ai/studio/case-studies): Explore case studies to see how Datasaur is proven to produce impact at the world's cutting-edge organizations.
- [Whitepapers | NLP Labeling](https://datasaur.ai/studio/whitepapers): A collection of papers featuring industry insights, data labeling information, and more. Read about the latest AI and NLP research with Datasaur's whitepapers.
- [Security | NLP Labeling](https://datasaur.ai/studio/security): Data Studio security: how Datasaur protects data during labeling projects.
- [AWS Partner | NLP Labeling](https://datasaur.ai/studio/aws-partner): Accelerate NLP projects with the power and scalability of the AWS cloud

## Company
- [Consensus Case Study](https://datasaur.ai/case-studies/consensus): Consensus case study.
- [About - Founder](https://datasaur.ai/about-founder): Founder background and the thesis behind Datasaur.
- [SLM Master Class Workshop: Learn Model Distillation for Efficient Language Models](https://datasaur.ai/webinar/slm-master-class-workshop): Join our free workshop on Dec. 12th at 1:00 PM ET to discover practical model distillation techniques for creating efficient Small Language Models (SLMs).
- [Register Now: SLM Master Class Workshop on Model Distillation](https://datasaur.ai/webinar/register-slm-model-distillation): Don’t miss this 35-minute workshop on Dec. 12th at 1:00 PM ET. Explore how to fine-tune Small Language Models (SLMs) from large LLMs. Hands-on learning with real-world applications.
- [Learn](https://datasaur.ai/learn): Learning material on LLMs, private deployment, and data workflows.
- [Resources](https://datasaur.ai/resources): Index of reports, guides, webinars, and whitepapers.
- [Our Mission for Private AI](https://datasaur.ai/about): Company mission: private AI for regulated industries.
- [Whitepapers](https://datasaur.ai/whitepapers): Long-form research on private AI, ownership, and deployment.
- [Guides](https://datasaur.ai/guides): Practical guides for evaluating and deploying private AI.
- [Case Studies](https://datasaur.ai/case-studies): Customer deployments and outcomes.

## Optional
- [The Leading NLP Labeling Tool for Audio and Text](https://datasaur.ai/nlp-labeling): Datasaur's original NLP labeling product page. See /studio for the current Data Studio product.
- [NLP Labeling Tool](https://datasaur.ai/nlp-labeling-tool): Datasaur's NLP labeling tool for text: a purpose-built editor, ML-assisted pre-labeling and team workflows for building NLP training data.
- [NER Labeling Tool](https://datasaur.ai/ner-labeling-tool): Named entity recognition labeling in Data Studio. Tag entities in text at scale with custom schemas, pre-labeling and reviewer workflows.
- [Datasaur OpenAI Integration](https://datasaur.ai/openai-integration): Use OpenAI models inside Datasaur Data Studio to pre-label text, then have your own team review and correct the output before training.
- [Natural Language Processing for AI](https://datasaur.ai/natural-language-processing-for-ai): NLP solutions from Datasaur: custom text models, advanced analytics and labeled training data built in Data Studio.
- [Weak Supervision NLP](https://datasaur.ai/weak-supervision-nlp): Weak supervision for NLP in Data Studio. Combine rules, heuristics and models to generate training labels at scale instead of annotating manually.
- [Data-Centric AI Solution](https://datasaur.ai/data-centric-ai): A data-centric approach to AI: improve model performance by improving training data. See how Datasaur Data Studio supports data-centric workflows.
- [Programmatic Labeling](https://datasaur.ai/programmatic-labeling): Automate annotation with data programming. Write labeling functions in Data Studio to label large NLP datasets without tagging every record by hand.
- [NLP Audio Labeling Tool](https://datasaur.ai/nlp-audio-labeling-tool): Label audio alongside transcripts in Data Studio. Datasaur's audio NLP tool supports transcription, span tagging and speaker-level review.
- [RLHF - Reinforcement Learning from Human Feedback](https://datasaur.ai/reinforcement-learning-human-feedback): RLHF tooling for LLM development. Collect human preference data, rank model responses and build reward datasets in Datasaur Data Studio.
- [NLP Sentiment Analysis Tool](https://datasaur.ai/sentiment-analysis-nlp): Build a custom sentiment analysis model on your own data. Label, train and evaluate sentiment classifiers in Datasaur Data Studio.
- [Datasaur | Productizing Large Language Models](https://datasaur.ai/productizing-large-language-models): Turning an LLM prototype into a production system: evaluation, deployment, and monitoring.
- [Data Labeling](https://datasaur.ai/data-labeling): Data Studio is Datasaur's text data-labeling tool. Label, review and manage NLP training data at scale with ML-assisted automation.
- [Text Labeling](https://datasaur.ai/text-labeling): Text labeling in Datasaur Data Studio. Build clean NLP training sets with an intuitive editor, automation and inter-annotator agreement checks.
- [Data Annotation](https://datasaur.ai/data-annotation): Annotate text for NLP in Datasaur Data Studio. Token, span and document-level annotation with review workflows and quality controls.
- [AWS Marketplace](https://datasaur.ai/aws-marketplace-datasaur): Datasaur on AWS Marketplace: private AI deployment and Data Studio labeling, procured through AWS.
- [Datasaur Terms of Service](https://datasaur.ai/terms-of-service): Terms of service.
- [Datasaur Privacy Policy](https://datasaur.ai/privacy-policy): Website privacy policy.