ChatBox AI
0%
Start for Free

04 / Enterprise Trust & Security

Security & Privacy Architecture.
Rigorous safeguards for confidential AI workflows.

An exhaustive breakdown of ChatBox AI's defense-in-depth model: contractual zero model training guarantees, in-transit and at-rest cryptographic isolation, hashed developer key management, and sovereign compliance under the Indian DPDP Act.

Zero Model Training TLS 1.3 & AES-256 PCI-DSS Level 1 Isolation Primary Hosting: India Developer: Arpit Ariyan Mahanarana (technon.co.in)DPDP Act & GDPR Aligned

Core Security Pillars

6 Architectural Safeguards
01Contractual Policy

Zero Public Model Training

Your prompts, code snippets, project briefs, and generated websites are never used to train, refine, or fine-tune public foundation models. We enforce strict enterprise zero-retention API contracts with OpenAI, Anthropic, and Google AI.

  • No prompt data leakage into public foundation weights
  • No evaluation or reinforcement training on customer inputs
  • Full intellectual property ownership over generated outputs
02End-to-End

Cryptographic Data Protection

All communications between your browser and our servers run strictly over TLS 1.3 with modern cipher suites. Database volumes, account identifiers, and conversation histories are encrypted at rest using industry-standard AES-256.

  • TLS 1.3 transport encryption with Strict Transport Security (HSTS)
  • AES-256 encrypted storage volumes across databases
  • Ephemeral memory processing for real-time model inferences
03Access Hardening

Multi-Factor Authentication & Identity

We support email-based One-Time Password (OTP) Multi-Factor Authentication to shield accounts against automated password stuffing and session hijack attempts.

  • MFA available directly in Account Settings
  • Strict JSON Web Token (JWT) session invalidation on logout
  • Automated rate-limiting and IP throttling against brute-force attacks
04Developer Safety

Hashed API Key Architecture

Developer API keys created for our OpenAI-compatible endpoint are stored strictly as salted SHA-256 hashes. Even in the event of an internal database audit, plaintext keys cannot be decrypted or stolen.

  • One-time display of secret keys upon initial creation
  • Salted SHA-256 cryptographic hashing in database tables
  • Instant one-click key revocation and rotation in Settings
05PCI-DSS Level 1

Payment & Financial Isolation

ChatBox AI servers never process, collect, or store raw credit card numbers, CVV codes, or net-banking passwords. All payment transactions are securely offloaded to certified payment partners.

  • Domestic Indian transactions (UPI, RuPay, Netbanking) via Razorpay
  • International subscription management via Paddle
  • Only tokenized transaction identifiers and invoice IDs are saved
06DPDP Act & GDPR

Data Sovereignty & Right to Erasure

In compliance with India’s Digital Personal Data Protection (DPDP) Act, users retain absolute sovereignty over their data. You can delete individual conversation threads or execute a full account purge at any time.

  • Primary database clusters hosted in Indian datacenter regions
  • Permanent wipe of profile records and threads within 30 days of deletion
  • Invoice records retained for 7 years per Indian corporate taxation laws

Security & Isolation Matrix

Across Tiers
Protection FeatureFree PlanPro Plan (₹499/mo)Max Plan (₹1,999/mo)
Zero Public Model TrainingIncludedIncludedIncluded
TLS 1.3 & AES-256 Volume EncryptionIncludedIncludedIncluded
Multi-Factor Authentication (MFA)IncludedIncludedIncluded
Developer API Keys (Hashed SHA-256)NoIncluded (60 req/min)Included (120 req/min)
Data Deletion & 30-Day Purge SLAIncludedIncludedIncluded
Support Response PriorityStandard (48–72h)Priority SLA (< 24h)Dedicated Engineering SLA

Sub-Processors & Data Sovereignty

Verified Partners

ChatBox AI strictly limits data transmission to vetted enterprise cloud providers. We never share, sell, or syndicate user data with data brokers, ad networks, or telemetry trackers.

Provider EntityRole & Data ProcessedDatacenter LocationCryptographic Standard
Firebase (Google)Authentication & Session Token VerificationIndia / GlobalEncrypted JWTs, OAuth 2.0
Supabase & AppwriteDatabase & Private File StorageIndia / FrankfurtAES-256 Volume Encryption, Row Level Security
Razorpay & PaddlePayment Processing & Subscription ManagementIndia / UKPCI-DSS Level 1 Certified, Tokenized Vaults
OpenAI, Anthropic, Google AIFoundation Model Inference ProvidersUS / EU / GlobalCommercial Zero-Training Enterprise APIs

Recommended User Security Practices

Shared Responsibility

Enable MFA

Turn on email OTP verification in Settings to prevent credential-stuffing attacks.

Protect API Keys

Never commit developer API keys into public Git repositories or client-side web bundles.

Sanitize Secrets

Avoid pasting production passwords, live tokens, or confidential PII into prompt text.

Purge Stale Threads

Regularly delete completed project conversations and website drafts from your dashboard.

Responsible Vulnerability Disclosure

Report Security Vulnerabilities

We welcome reports from independent security researchers. If you identify a potential security vulnerability, report it to our Data Protection Officer and security engineering team for prompt investigation and triage.

Contact DPO Full Privacy Policy
ChatBox AI↑

The Ultimate Multi-Model AI Workspace. Chat, research, generate code, create images, and deploy web applications with 56+ frontier models.

Explore

AI ModelsAndroid AppDeep ResearchSecurity & PrivacyPricing Plans

Solutions

For DevelopersFor ResearchersFor CreatorsWebsite Builder

Support

Help & SupportDocumentation ↗Model API DocsStart Free Account

Legal & Trust

Privacy PolicyTerms & ConditionsData ProtectionCookie Choices

© 2026 ChatBoxAI · Developed by Arpit Ariyan Mahanarana (Technon). All rights reserved.

Privacy Policy·Terms & Conditions·Help & Support·Back to top ↑
ChatBox AI

Cookie Preferences

Consent

We use strictly necessary storage for authentication and optional privacy-first analytics to improve ChatBoxAI. No personal data is ever sold.

Read our Privacy Policy