AI Innovation Centre
AI Cyber Safety Assistant
এআই সাইবার নিরাপত্তা সহায়ক
Paste the message, link or profile description you are unsure about. You get the risk indicators, why each one matters and what to do next — never an accusation against a named person.
Safe identity tools — the exact limits
Phone, photo and document tools run only inside lawful, consented boundaries. These limits are published so anyone can hold the Forum to them.
Phone number tools ফোন নম্বর টুল
- ✓ Look up your own account after an OTP sign-in
- ✓ Verify an NFF member or an official Forum contact number
- ✓ Validate a published business or public contact number
- ✓ Check a number against reported scam-number signals
- ✓ Duplicate-account checks under authorised internal rules
- ✕ Revealing private personal information behind an arbitrary number
- ✕ Reverse person lookup without consent or lawful authority
- ✕ Any form of doxxing or hidden-data aggregation
Photo & image tools ছবি ও ইমেজ টুল
- ✓ OCR — reading text out of a document photo
- ✓ Document photo quality and completeness checks
- ✓ Image safety screening and misuse reporting
- ✓ Duplicate image and provenance indicators
- ✓ Manipulated-content indicators (advisory)
- ✓ Consented identity verification against NFF records
- ✕ Identifying an unknown person from a face
- ✕ Searching private identities from a photo
- ✕ Any face-based stalking or surveillance use
ID & document tools আইডি ও ডকুমেন্ট টুল
- ✓ OCR and field extraction from a document you own
- ✓ Format and completeness validation
- ✓ NFF-issued ID and certificate verification by serial or QR
- ✓ Consented verification inside authorised records
- ✓ Fraud-risk indicators for a human officer to review
- ✕ Validating a national identity document as genuine on NFF authority
- ✕ Storing or displaying full identity numbers anywhere public
Verify an NFF card or certificate at the verification portal, or open a tracked case at the Cyber Command Center.
Responsible AI notice
দায়িত্বশীল এআই ঘোষণা
Privacy. Personal identifiers in what you submit are masked before the request reaches a model. We do not sell your data and we do not use your submissions to train external models.
Limits. These tools give assistance, not certainty. Output can be incomplete or wrong, and it is not legal, medical or financial advice. Always verify against the original source.
Human review. Any consequential conclusion — a fact-check verdict, a safety or moderation decision, a complaint outcome — is reviewed by an NFF officer before it is acted on.
Safety. Instructions inside uploaded files or pasted text are treated as content, never as commands. We do not provide attack or intrusion guidance, and serious harm is referred to lawful channels.
Full limits: what our AI can and cannot do · report a problem to a human
Other AI tools
AI Tutor
এআই শিক্ষক
Lesson explanation, study plan, practice questions and weak-topic detection.
Career AI
ক্যারিয়ার এআই
CV, portfolio and proposal improvement with a real skill-gap analysis.
Fact Check AI
ফ্যাক্ট চেক এআই
Breaks a viral claim into checkable parts, then hands it to a human reviewer.
AI Fact-Check & Media Verification
ফ্যাক্ট-চেক ও মিডিয়া যাচাই
Photo, video, audio and news-text checkers with a combined verification report and specialist review.
Read the limits first: what AI can and cannot do.

