NT (No Threats)

The description states that NT (No Threats) built SilverGuard, a digital gatekeeper for elderly users, using GPT-Live and telecom infrastructure to detect fraud in real-time and involve family members in calls when suspicious activity is detected. The system integrates VoIP/SIP trunking with OpenAI’s GPT-Live API and uses function calling to initiate conference calls with family members.

The project is self-reported as a hackathon submission, not independently verified. No evidence of revenue, customers, traction or commercialization exists in the description. It appears to be an experimental prototype built for demonstration purposes.

The description states that SilverGuard is a "digital gatekeeper" for elderly users, designed to intercept incoming calls and analyze them in real-time using GPT-Live. If suspicious activity is detected, it triggers an automated call to family members via function calling.

• It uses VoIP/SIP trunking to intercept incoming calls.
• It leverages OpenAI’s GPT-Live (realtime API) for voice-to-voice interaction with low latency.
• It uses OpenAI Function Calling to initiate conference calls with family members when fraud risk is detected.

The system does not appear to be a consumer-facing product but rather an experimental prototype built for a hackathon.

Inference The product appears to be a proof-of-concept, not a commercial offering. It is described as a "Tech Demo" and lacks evidence of real-world deployment or user adoption.

The description states that SilverGuard aims to protect elderly users from fraud by acting as a digital gatekeeper. It positions itself as an AI-powered solution that can detect deception in real-time and involve family members in decision-making during suspicious calls.

It claims to:

• Replace human intervention with AI for fraud detection.
• Use GPT-Live for real-time voice interaction.
• Enable family involvement through automated conference calling.
• Be scalable to telecom network-level services.

Inference The positioning is centered on protecting vulnerable users, but the description does not indicate any commercial traction or adoption beyond the hackathon context. The claims are aspirational and unverified.

The description states that SilverGuard targets elderly users who are vulnerable to fraud from scam callers. It aims to protect those who may be isolated or unable to assess suspicious calls on their own.

It also mentions:

• Family members as a secondary user group.
• Telecom providers as potential partners for network-level deployment.

Inference The primary customer is the elderly, and the ICP appears to be individuals aged 60+ who are at risk of fraud. No evidence of actual customers or market segmentation exists in the description.

The description does not provide any information about pricing, monetization or business model.

It mentions:

• Potential telecom partnerships.
• Optional service for elderly users.
• Integration with financial APIs (not described as a paid feature).

Inference There is no evidence of a defined business model or pricing structure. The project appears to be experimental and not yet commercialized.

The description states that the system was built using:

• GPT-Live (OpenAI’s real-time API)
• VoIP/SIP trunking
• Function calling for dynamic action routing
• Next.js for frontend
• WebRTC for voice communication

It also mentions:

• Challenges in achieving ultra-low latency.
• System prompt engineering to avoid false alarms and maintain natural conversation.

Inference The technical stack is described as advanced, but there is no evidence of production-grade delivery or scalability. It’s a hackathon prototype.

The description states that this is a hackathon project submitted to the OpenAI 2026 hackathon on Devpost.

It mentions:

• The team built a working prototype.
• It was able to detect fraud scripts and trigger family involvement.
• It is not yet deployed or commercialized.

Inference No traction, revenue, or adoption data are provided. The project is described as a demo, not a product in the market.

The description does not mention any competitors or existing solutions in the fraud detection or elderly protection space.

It implies that there is a gap in the market for AI-powered fraud detection integrated with telecom infrastructure and family involvement.

Inference No competitive analysis is provided. The project appears to be positioned as an innovative concept, but no evidence of prior or competing products exists.

• Unverified claims: All features and functionality are self-reported.
• No traction or commercialization: The system has not been deployed or tested beyond a hackathon.
• Technical feasibility concerns: Ultra-low latency in real-time voice AI is challenging, especially with telecom integration.
• Privacy and ethics risks: Involving family members in calls raises privacy concerns.
• Scalability assumptions: The project is described as a prototype, not a scalable solution.

Inference The project lacks evidence of real-world testing or commercial viability. Risks are high due to the experimental nature and lack of independent validation.

1. What specific fraud patterns does the AI detect? How is it trained?
2. Has the system been tested with real users or in a controlled environment?
3. What are the technical limitations of GPT-Live in real-time voice interaction?
4. How does the system handle false positives and avoid blocking legitimate calls?
5. Are there any privacy or legal concerns with involving family members in calls?
6. Is there any plan for telecom partnerships or commercial deployment beyond the hackathon?
7. What is the current status of the prototype? Is it being tested or used by anyone?