Shield AI — All Plans

Your Always-On DPDPA Compliance Co-Pilot

Ask compliance questions in plain English. Get DPDPA section-cited answers grounded in your organisation's live data. Draft policies. Classify PII. All without sharing personal data with the AI.

What Shield AI can do
Answer DPDPA questions
Section-cited answers grounded in your compliance state
Draft privacy policies
Pulls RoPA + purposes + processors into a structured draft
Classify PII
Special Category / Sensitive / Personal — with data categories
Surface priority prompts
Context-aware suggestions from your top failing controls

PII-free by design. Only compliance metadata is shared with the model — never personal data, email addresses, or consent records.

From question to DPDPA-grounded answer

1
Org context injected
Before every query, Shield AI reads your live compliance score, active controls, top failing modules, and registered processors — all as plain-text metadata. No PII leaves your tenant.
2
Ask anything
Type a compliance question in plain English. Shield AI cites the exact DPDPA section number in every answer so you can trace its reasoning back to the Act.
3
Classify or draft
Paste an asset description to get an instant PII classification (Special Category / Sensitive / Personal / Anonymised) with a confidence score and data category suggestions.
4
Generate policy draft
One click pulls your RoPA, purposes, processors, and retention schedules into a structured DPDPA-compliant privacy policy draft — ready to edit and publish.
5
Suggested prompts
Shield AI surfaces context-aware prompts based on your top failing controls — so you always know the highest-priority question to ask next.

Compliance intelligence, always available

Core

Org-Aware Context

Compliance score, module health, deadlines, active incidents, and processor list injected as context on every query. Answers are specific to your organisation — not generic.

DPDPA Section Citations

Every answer includes the exact Act section and sub-section number. No hallucinated legal advice — traceable citations every time.

PII Classifier

Paste any asset description, data field list, or form schema. Get classification level, confidence, suggested data categories, and DPDPA-specific handling notes.

Privacy Policy Auto-Draft

Pulls your RoPA data, processing purposes, processors, and retention policies to generate a structured DPDPA 2023 compliant privacy policy with correct Section 5 disclosures.

Suggested Prompts

Context-aware prompt suggestions based on your top 3 failing DPDPA controls — never stare at a blank chat box wondering what to ask.

Privacy

Zero PII in Prompts

Shield AI is architected to never include personal data in prompts. Only aggregated metadata, scores, and counts are shared with the model. DPDPA-safe by design.

All
Plans included
0 PII
In prompts
5K
Enterprise quota/mo
<5s
Response time
Available from Starter plan onwards

Shield AI is available on all plans. Starter: 50 msgs/mo. Growth: 200. Business: 1,000. Enterprise: 5,000 + ₹10/msg overage.

Shield AI — answered

Does Shield AI ever see my users' personal data?+

No. Shield AI is architected to be PII-free by design. The context injected into every prompt is aggregate metadata only: compliance scores, module health indicators, counts (e.g. "12 open rights requests"), processor names, and control statuses. No email addresses, consent records, personal identifiers, or individual data records are included. This is enforced at the service layer — not a configuration option.

Which AI model powers Shield AI?+

Shield AI uses Claude (Anthropic) via the Claude Sonnet model. We chose Claude because it follows instructions reliably, keeps answers grounded in provided context, and is less prone to hallucinating legal citations than other models. The model is instructed to cite DPDPA section numbers in every compliance answer.

How accurate are the DPDPA section citations?+

Shield AI is instructed to cite section numbers from the DPDPA 2023 and DPDPA Rules 2025 that are provided in its system prompt. It will not invent citations. If it cannot find a grounding section, it will say so. You should treat citations as a starting point for your DPO's review — not as final legal advice.

What is the monthly message quota?+

Starter: 50 messages/month. Growth: 200/month. Business: 1,000/month. Enterprise: 5,000/month, then ₹10/message overage (AI keeps working, never blocks). Quotas reset on the 1st of each month. Super admins can view and manage overage in the admin panel.

Can I use Shield AI to draft contracts or respond to regulators?+

Shield AI is a compliance co-pilot, not a legal firm. It's excellent for drafting privacy policy language, classifying data, and explaining your obligations — but any output used in regulatory submissions or contracts should be reviewed by your legal team or a qualified DPO before use.

Ask your first compliance question today.

Shield AI is included on every plan — no extra setup needed. Start a free trial and get an answer in under 5 seconds.