About TRACE Shield
What it is, why it exists, and how it works
A plain account of the product, written for anyone deciding whether to trust it, not just for the person about to use it.
Why it exists
Most modern scams don’t defeat technology first, they manipulate a real person into authorising the harmful action themselves. A bank can detect a stolen card well. It is far harder to stop a transfer when the customer enters the OTP themselves, approves the UPI request themselves, sends the bank transfer themselves, or installs the remote-access app themselves.
This is not victimless. It causes direct financial loss, fear, shame, debt, damaged trust, and family stress, sometimes severe emotional harm. Anyone can be targeted. Vulnerability here is created by pressure in the moment, not by a lack of intelligence.
The financial cost is large and current: official 2024-2025 data puts losses at roughly ₹22,845.73 crore in India, £1.28 billion in the United Kingdom, and $2.18 billion in Australia, see the sources on the homepage. The harder cost is the moment itself, a message or call designed to create panic, arriving when there is no time to think and no one to ask. Most existing advice is generic (“be careful of scams”) or arrives only after the money is gone. TRACE Shield is built for that specific moment, the pause between the scammer’s request and the click, the OTP share, the app install, or the payment.
Which scams cause the most harm, by country
Ranked by reported financial loss, not by how many reports were filed, a small number of scam types cause most of the damage. Investment and trading scams top the list in every market TRACE Shield covers.
India
- 1Stock trading scams: ₹4,636cr, 2.28 lakh complaints
- 2Investment scams: ₹3,216cr, 1.00 lakh complaints
- 3‘Digital arrest’ fraud: ₹1,616cr, 63,481 complaints
Jan-Sep 2024, I4C data via Ministry of Home Affairs.
United Kingdom
- 1Investment fraud: £221.5m, up 40%
- 2Purchase scams: £118.1m, 71% of APP cases
- 3Romance fraud: £39.2m, up 23%
2025, UK Finance Annual Fraud Report 2026, APP fraud only.
Australia
- 1Investment scams: $837.7m
- 2Payment redirection scams: $166.8m
- 3Romance scams: $139.9m
2025, ACCC / National Anti-Scam Centre, Targeting Scams report.
Why losses continue
Scammers attack people, not only systems
A scammer builds an urgent story: your account will be blocked today, there is an arrest warrant, your child is in trouble, you’ve won an investment opportunity, your parcel is held, a supplier changed their bank details, don’t tell anyone and transfer now. This is social engineering: rushing, frightening, or isolating someone, or promising a reward, so that even a normally cautious person makes a decision they’d otherwise question.
In the UK, authorised-push-payment fraud, where the victim is deceived into sending the money themselves, caused £576.4 million in losses in 2025 across 248,070 recorded cases. UK Finance reports that 66% of these cases began online and 17% began via telecommunications, but that telecom-originated scams accounted for 28% of the losses, a disproportionate share that’s a large part of why call protection matters here, not just message and link checking.
Payments are instant, certainty is slow
UPI, Faster Payments, PayID, cards, and bank transfers make legitimate payments fast and convenient, but once a scam payment is approved, funds can move through several accounts within minutes. The realisation often comes later: a fake authority was trusted, an OTP or PIN was shared, a “collect” request was approved believing money would be received, or a transfer went to a “safe account.” By the time a bank or investigator gets involved, the money may already be withdrawn, converted, or layered through mule accounts.
Where TRACE Shield fits
Banks, spam filters, browser warnings, and police all play essential roles, and none of them is being replaced. Most work automatically before a scam reaches you, or step in afterwards if money is already gone. TRACE Shield is built for the moment in between: the pause after a suspicious message or call arrives, and before you act on it.
Before it reaches you
Bank fraud systems, spam filters, browser warnings, caller-ID apps
Block known threats automatically, no action needed from you.
The moment it arrives
TRACE Shield
A pause, a plain-language explanation, and a way to check with someone you trust, before you act.
After, if money is already gone
Banks, police, and cybercrime reporting portals
Investigate, freeze transactions, and coordinate a response.
What TRACE Shield actually changes
In that moment, TRACE Shield gives a person four things, quickly.
A pause
A clear interruption: stop, do not pay or share anything yet.
An explanation
Named, specific reasons, for example: this caller claims arrest, demands secrecy, and asks for money, all known impersonation-scam signals.
A human check
A Safety Circle: send this to someone you trust before transferring money, especially valuable for teenagers, isolated users, and older adults.
The correct response
If harm may already be underway: contact the bank or provider, preserve the transaction reference and screenshots, and use the right local reporting route.
Australia’s scam-prevention model is built around a related principle: scam disruption should prevent contact, stop contact once it’s happened, and prevent the payment. TRACE Shield applies that same logic directly to an individual’s decision point.
What it is
A six-layer human-AI safety system, modelled on established risk-management structure (NIST Cybersecurity Framework’s Govern, Identify, Protect, Detect, Respond, Recover; Australia’s Scams Prevention Framework’s prevention, detection, disruption, response, and reporting), adapted for an individual facing a scam in real time rather than an institution managing risk on a quarterly cycle.
Identify
Understand the country, channel, payment type, and how far things have already gone.
Protect
Stop risky actions before you click, share, install, or pay, shown immediately based on what you're dealing with, before any analysis even runs.
Detect
Analyse the message, link, sender, or a suspicious call (Call Shield) for explainable scam signals, OTP requests, urgency language, lookalike domains, impersonation patterns like 'digital arrest', never a black-box score.
Verify
For high-risk cases, bring in independent verification and a trusted person, your Safety Circle, before you act.
Respond
If harm may already be underway, get a clear, immediate action plan and the right local reporting route for your country.
Learn & Govern
Privacy, human oversight, and an honest account of what the system does and doesn't do, published openly, not buried in terms of service.
How it works
This is not an AI chatbot producing unsupported claims. It is a deterministic rule engine, plus a small number of explicit override rules for known-dangerous combinations (an OTP request plus urgency, for example), designed so every result is traceable and defensible, not guessed.
Explainable rules
Every risk level traces back to named, visible signals, not a hidden score. The “Why” shown for each check is the actual reasoning, not a summary of it.
Human judgement stays central
AI identifies and explains risk. You, or someone in your Safety Circle, make the consequential decision, TRACE Shield never acts, pays, or contacts anyone on its own.
Nothing leaves your browser
The check runs entirely on the page you’re using, deterministic rules, no external AI call, no message content sent anywhere, no exceptions.
Country-specific routes
Real reporting channels for India, the UK, and Australia (1930, Report Fraud, Scamwatch and equivalents), not generic advice to “contact the authorities.”
The full commitments and an honest account of what’s built versus designed for later are on the privacy & governance page.
Who it protects
No single age group should be stereotyped as “the victim.” In Australia’s 2025 data, older Australians reported the highest total losses, but people aged 25-44 were the most likely to report a financial loss at all.
| User group | Common risk | TRACE Shield value |
|---|---|---|
| Teenagers | Sextortion, gaming and in-game item scams, fake jobs, social-media impersonation, giveaways. | A non-judgemental route: don't pay, save evidence, tell a trusted adult. |
| Young adults | Job, rental, shopping, loan, investment, and marketplace scams. | Link and message checks plus a payment-pause warning before money moves. |
| Working adults | Bank impersonation, delivery, tax, investment, remote-access, and business-payment scams. | Fast verification and a clear response plan if something's already happened. |
| Seniors | Police, bank, and government impersonation, romance, investment, and remote-access scams. | Clear warnings, a Safety Circle, and simple emergency steps under pressure. |
| Families | A loved one being pressured into paying while isolated from anyone who could sanity-check it. | A consent-based trusted-person review before the payment goes ahead. |
| Small businesses | Fake invoices, changed supplier bank details, CEO impersonation. | A verify-before-pay workflow and a record of what was checked. |
Why it is not “victimless”
A scam can look like “just a bank transfer,” but the real impact often includes life savings or emergency funds lost, debt taken on trying to “recover” a loss, lost rent, school fees, medical funds, or payroll, trauma and embarrassment that makes people reluctant to seek help, family conflict and a lasting loss of confidence in digital banking, and re-victimisation, since scammers often target the same person again.
Investment scams are particularly damaging because victims can be persuaded over weeks or months into repeated payments. UK Finance reported £221.5 million in investment-fraud losses in 2025, up 40% year-on-year.
How success is measured
TRACE Shield does not promise to catch every scam. It promises a clear pause, understandable risk evidence, trusted human support, and immediate local action, before or after a scam. Success is measured by outcomes, not by a detection rate alone:
- Fewer people clicking high-risk links.
- More people refusing to share OTPs, PINs, or passwords under pressure.
- More high-risk payments independently verified before they're sent.
- Faster contact with the bank after a loss.
- Better evidence records for bank and police reports.
- Better decisions in scam simulations.
- Lower loss severity among people who use the emergency-response flow.
Where it came from
TRACE Shield extends a body of financial-crime research and tooling built by Himaja Prabhala, who previously worked as a financial crime analyst and has spent the past year researching and building TRACE, focused on efficient systems and AI explainability. The research paper, TRACE: An Interactive AI and Compliance-by-Design Framework for Financial Crime Detection, is documented, citable, and open to review at doi.org/10.5281/zenodo.20699574.
The first system built from that research was TRACE Risk Advisory’s compliance engine. TRACE Nexus followed as a larger human-AI workspace for professional investigators, built on the same direction. TRACE Shield is the newest piece: the same underlying approach, explainable rules and human judgement over black-box AI, adapted for direct, everyday use, so anyone facing a scam themselves gets the same explainable protection a professional investigator would.