Archive position — measured, not model output
0 likes on Devpost
2,264 of the 7,856 archived projects have more likes, and 5,592 share exactly 0 — so this project's #6,552 place in the like-ranked listing is a tie-break inside that group, not a ranking.
Projects (log scale)
Likes on Devpost. ▲ marks this project's group.
Show the figures
| Likes | Projects | Share of archive |
|---|---|---|
| 0 | 5,592 | 71.2% |
| 1 | 1,758 | 22.4% |
| 2 | 285 | 3.6% |
| 3–4 | 132 | 1.7% |
| 5–9 | 75 | 1.0% |
| 10+ | 14 | 0.2% |
Executive Summary
What the company appears to be
ScamLens is an AI-powered fraud prevention tool designed for everyday users in India. It allows users to paste messages, upload screenshots, or forward WhatsApp messages to detect scams and provide actionable guidance. The system uses a deterministic and LLM-based approach with a focus on precision over recall, and it enforces strict honesty constraints (e.g., no "safe" verdicts). It also includes a community intelligence layer called Sentinel that aggregates scam patterns.
What changed
The project is a self-reported hackathon submission. No evidence of prior traction, revenue, or customer adoption exists beyond the author’s own description. The product is described as built in a short timeframe for a hackathon, with no indication of commercial deployment or scale.
Single most important open question
Is there any evidence that ScamLens has been adopted by users or integrated into real-world systems, or is it purely a prototype?
What The Product Actually Is
The description states that ScamLens is an AI fraud prevention agent for everyday users. It allows users to input scam-related messages or screenshots via WhatsApp or web interface and returns:
- A risk verdict with specific manipulation tactics
- Playbook name (e.g., digital arrest, UPI collect-request)
- Immediate action the user can take
- Official guidance routed to real authorities
It also includes a community intelligence layer called Sentinel that shows active scam families.
Evidence
- “Paste a message, upload a screenshot, or forward it to ScamLens on WhatsApp.”
- “In seconds you get: A risk verdict with the specific manipulation tactics found, not a confidence score”
- “Every scan also feeds Sentinel, a live community intelligence layer showing which scam families are active”
Inference The system uses a multi-agent architecture (LangGraph) and integrates vision API for screenshot input.
Positioning & Claim Evolution
ScamLens positions itself as a zero-friction AI agent that detects fraud in seconds, with no false all-clears. It is built to address the gap between doubt and action — users often feel uncertain but don’t act because checking is time-consuming or embarrassing.
Evidence
- “ScamLens is a zero-friction AI agent that detects fraud in seconds”
- “A bot is not a human. It has no opinion of you and tells no one.”
- “ScamLens never says 'safe.' That rule is enforced in code — app/copy.ts throws a build-time error if the words 'safe,' 'not a scam,' or 'safe to pay' appear in any user-facing string.”
Inference The positioning reflects an attempt to solve a behavioral problem (delayed action) rather than just a technical one (detection).
Target Customer & ICP
The target customer is everyday users in India who are vulnerable to cyber scams, particularly those involving WhatsApp messages or digital arrest scripts.
Evidence
- “ScamLens is a zero-friction AI agent that detects fraud in seconds, shows the pattern behind it, and feeds every catch into a community intelligence layer.”
- “India loses $2.7B a year to cyber scams.”
Inference The ICP appears to be individuals who receive scam messages via WhatsApp or other digital channels and are likely to be misled by social engineering tactics.
Business Model & Pricing Evidence
There is no evidence of pricing, monetization strategy, or business model in the description. The project is described as a hackathon submission with no indication of commercial viability or revenue streams.
Evidence
- No mention of pricing tiers, subscriptions, or monetization
- No evidence of customer acquisition or retention strategies
Inference The product may be intended for public good or early-stage testing, but no business model is evident.
Technical & Delivery Signals
The system is built using Next.js, React, TypeScript, LangGraph, and integrates with OpenAI APIs. It uses deterministic signal detection alongside LLMs, has a privacy node that strips PII before LLM processing, and includes an evidence gate to ensure all verdicts are justified.
Evidence
- “Next.js 16 + React 19 + TypeScript, with a LangGraph state machine doing the actual reasoning.”
- “Privacy node runs before any LLM call. Aadhaar numbers, phone numbers, OTPs, emails and UPI IDs are stripped from the text before it leaves the process.”
- “An evidence gate is the final node. No verdict ships without cited reasons attached.”
Inference The architecture emphasizes safety, traceability, and deterministic fallbacks — signals of a thoughtful engineering approach.
Traction & Maturity Signals
There is no evidence of traction or maturity beyond the hackathon submission. No customers, revenue, or adoption data are provided.
Evidence
- “This project was submitted to the OpenAI 2026 hackathon on Devpost.”
- “No revenue, customer or traction data is available beyond what they state.”
Inference The product is in a prototype or early-stage development phase and has not yet been deployed at scale.
Competitive Context
There is no evidence of direct competitors or competitive positioning. The description does not mention existing tools for fraud detection or scam prevention, nor does it indicate how ScamLens would differentiate from them.
Evidence
- No mention of competitors
- No indication of market analysis or differentiation strategy
Inference The competitive landscape is unknown; the product may be addressing an underserved niche or a new category.
Key Risks & Red Flags
- No commercial traction or adoption: The project is described as a hackathon submission with no evidence of real-world use.
- Unproven scalability: No evidence of infrastructure, user base, or performance at scale.
- Limited data validation: The team used real-world cases for testing but did not provide metrics on how many users are currently using the tool.
- No monetization strategy: No indication of how the product will generate revenue.
- Over-reliance on deterministic fallbacks: While this is a strength, it may limit the system’s ability to evolve with new scam patterns.
Evidence
- “Everything above is the authors' own account. It is not independently verified.”
- “No revenue, customer or traction data is available beyond what they state.”
Diligence Questions To Ask The Founders
- What is the current status of ScamLens? Is it being used by individuals or organizations?
- How does the team plan to scale the community intelligence layer (Sentinel)?
- Are there any partnerships with telecom providers, banks, or government agencies?
- What are the long-term plans for monetization and commercial viability?
- How is the system trained on new scam types, and how often is it updated?
- What is the team’s plan to expand beyond India and into other markets?
Investment/Partnership Verdict
Not evidenced.
There is no evidence of revenue, customers, or traction to support an investment or partnership decision. The project is described as a hackathon submission with no indication of commercial viability or scalability. While the idea and execution show promise, there is insufficient data to assess its potential for growth or return.
Confidence Low — based entirely on self-reported description, with no external validation or evidence of adoption.
Source
Submitted to the OpenAI 2026 hackathon on Devpost. Project home on DevPost.
The analysis above was generated by a language model from the project's own one-line description. It is not independent research and contains no verified traction, revenue or customer data.