Local
Private
AI
Local,
private
AI
What happens now to data used in public AI?
-
01 User submits chats, queries, docs into web based public AI model
Once the data is submitted, it’s visible to the public AI model owner — it must be visible for processing. Data transmitted across the internet isn’t inherently visible, however, it does mean there are more third parties involved and there’s more chance of interception.
-
02 The public AI infra machine
Most public AI companies use data submitted by users to “improve their services” and, if not explicitly requested, they use for “training purposes”. The public AI infra stack normally includes their: backend OS; LLM; model training; product improvement; and data storage. Note that most public AI stores user data which creates more data breach risk.
-
03 Third parties process much of users' data
Most public AI companies send user data to many third party processes. E.g., OpenAI uses Microsoft, Cloudflare, CoreWeave, Oracle, Google, Snowflake, Salesforce and many more, and in numerous countries. OpenAI’s dynamic Sub-Processor List is here.
-
04 The upshot
User and company data, personal and confidential info, and IP are shared with many unknown third parties in different countries for commercial reasons that don’t benefit the user. This increases company costs and may lead to possible harm later from data misuse.
What happens to data with Zainode’s AI on the client’s infra (DC or on-prem)?
-
01 User submits chats, queries, docs into Zainode's web app
Once the data is submitted, it’s visible within the client’s infra, only to the client and who they delegate access to. Client is in complete control of how their users connect to Zainode app, e.g. VPNs, offline completely, or other strong mechanisms of private transmission.
-
02 Client's private infra, provided by Zainode
Zainode provides intake web app, LLM, servers and data storage. Note that Zainode’s data storage app does store data, but only in the client’s infra — the client is in full control of access, retention, deletion.
-
03 No third parties at all
None. Zero. Zainode’s private AI solution uses no third party data processors.
-
04 The upshot
User and company data, personal and confidential info, and IP are stored by the client, in the client’s complete control. This considerably mitigates data misuse and reduces tech costs.
Unlike most public AI, all AI generated output by Zainode is provided with evidence and reasoning.
What happens to data with Zainode’s AI on Zainode infra?
-
01 User submits chats, queries, docs into Zainode's web app
Once the data is submitted, it’s visible only to the client and who they delegate access to. Zainode can connect directly to a client’s network to minimize parties involved in data transmission and to client authentication system so users are centrally managed. Client has full log of all connections and use.
-
02 Zainode's infra, logically & physically isolated per client
Zainode provides intake web app, LLM, servers and data storage. Note that Zainode’s data storage app does store data, but the client is in full control of access, retention, deletion. No tenancy is shared between clients. Full data logging available.
-
03 No third parties at all
None. Zero. Zainode’s private AI solution uses no third party data processors.
-
04 The upshot
User and company data, personal and confidential info, and IP are stored by us in the client’s single tenancy, in your complete control. This considerably mitigates data misuse and reduces tech costs.
Unlike most public AI, all AI generated output by Zainode is provided with evidence and reasoning.
Your model, your data, your AI
-
Each client accesses single-tenant application, with fully customizable interface showing your brand. Your staff can use your AI model with your company data.
-
Know everything in and out of AI, including AI’s reasoning, enabling better explanation of AI’s decisions and advice, and liability and risk management. Prompts and outputs stay entirely within your environment — we don’t access the data you process.
-
Zai Node’s infrastructure handles the computing demands of running modern-day AI inference workloads, helping bring your innovation to market privately and faster. Clients access the latest GPUs for ultra-high performance, low cost and ease of use. Your data is transiently processed by your GPUs.
-
Safely and securely unlock the benefits of AI. Organizations who keep AI closed to their data unlock powerful insights and competitive advantages — all while maintaining trust and compliance.
-
Zai Node avoids the expense, cost uncertainty, wastage and lack of visibility with the public cloud.
Your model, your data, your AI
Custom prompt application
Each client accesses single-tenant application, with fully customizable interface showing your brand. Your staff can use your AI model with your company data.
-
Know everything in and out of AI, including AI’s reasoning, enabling better explanation of AI’s decisions and advice, and liability and risk management. Prompts and outputs stay entirely within your environment — we don’t access the data you process.
Purpose built infra, not cloud
Zai Node’s infrastructure handles the computing demands of running modern-day AI inference workloads, helping bring your innovation to market privately and faster. Clients access the latest GPUs for ultra-high performance, low cost and ease of use. Your data is transiently processed by your GPUs.
Trust & compliance
Safely and securely unlock the benefits of AI. Organizations who keep AI closed to their data unlock powerful insights and competitive advantages — all while maintaining trust and compliance.
-
Zai Node avoids the expense, cost uncertainty, wastage and lack of visibility with the public cloud.
| Feature | On-prem | Our datacenter | Your datacenter | |
|---|---|---|---|---|
| Data control | Full data control | Managed data security; we never store model inputs or outputs | Full data control in your DC; managed data security | |
| Data residency requirements | Contained in country | Multi-region support with global deployment options | Region-locked data and deployments with multi-region support | |
| Compute capacity | Customized to order | Customized to order | Leverage existing resources or Zai Node compute for overflow | |
| Cost efficiency | High | Cost-effective, on-demand compute | Use in-house compute whenever available for optimized costs | |
| Cost certainty | Fixed | Fixed and flexible | Fixed and flexible | |
| Integration with internal systems | Custom or out-of-the-box integrations | Easy integration via ecosystem | Custom or out-of-the-box integrations | |
| Performance optimization | On-chip model performance | On-chip model performance and low network latency | On-chip model performance and low network latency | |
| Scalability | High, tailored scalability | High, flexible scaling options | High, flexible scaling options | |
| Security and compliance | Organizational policies | SOC 2 Type II, HIPAA, and GDPR compliant | Organizational policies | |
| Support and Maintenance | Comprehensive support and managed services | Comprehensive support and managed services | Comprehensive support and managed services | |
| Utilization of existing cloud commits | Spend down existing | Spend down existing | Use credits or commits | |