DeshkaAI – Human-First AI Safety Infrastructure

DeshkaAI is a self-reported AI safety infrastructure prototype built for high-impact environments such as emergency response, cybersecurity, smart cities, healthcare, and public infrastructure. The project was submitted to the OpenAI 2026 hackathon by a single founder, Arshad Khan. It claims to introduce safety layers before AI actions, including context verification, risk assessment, stability checks, and human oversight.

The description states that DeshkaAI is a Human-First AI safety infrastructure that verifies context, assesses risk, applies stability checks, and involves human oversight before critical actions. The system integrates modern AI models for reasoning and includes independent safety verification before responses are released.

There is no evidence of revenue, customers, or traction beyond the prototype's existence. The project is presented as a proof-of-concept with no indication of commercial deployment or adoption.

The description states that DeshkaAI V12 is a Human-First AI safety infrastructure. It aims to help AI systems verify context, assess risk, apply stability checks, and involve human oversight before critical actions.

It includes the following components:

• Context Verification
• Risk Assessment
• Stability Checks
• Human-First Safety Gate
• Safe Action / Safe Hold
• Decision Logging

The system is built using Python with modular safety components. It integrates modern AI models for reasoning while adding independent safety verification before responses are released.

Inference The product appears to be a prototype or proof-of-concept, not a commercial offering. It is described as a "Human-First" infrastructure, suggesting it is designed to maintain human control in high-stakes AI decision-making.

The description states that DeshkaAI was created to explore a Human-First AI safety infrastructure that verifies context before action. The goal is to improve the reliability of AI systems operating in critical environments such as emergency response, cybersecurity, smart cities, healthcare, and public infrastructure.

It positions itself as a solution for AI systems that are becoming more autonomous but need to be safe in high-impact environments.

There is no evidence of prior positioning or evolution of claims beyond this single submission. The project is described as a hackathon prototype, not a product with a history of development or market positioning.

The description states that DeshkaAI is intended for AI systems operating in critical environments such as:

• Emergency response
• Cybersecurity
• Smart cities
• Healthcare
• Public infrastructure

These are high-stakes domains where incorrect decisions can have serious consequences, and the system aims to improve reliability through safety layers.

There is no evidence of specific customer segments or personas beyond these general categories. The description does not indicate whether the product targets enterprises, government agencies, or developers building AI systems.

The description does not provide any information about pricing, monetization, or business model. It is a self-reported prototype submitted to a hackathon and lacks evidence of revenue streams, customer contracts, or pricing structures.

Inference The product is likely in early development and has no commercialized business model at this stage.

The project is built using:

• Python
• Modular safety components
• Modern AI models (e.g., codex, gemini-api, openai-gpt-5.6)
• REST API integration
• HTML, CSS, JavaScript, JSON

It integrates AI models for reasoning and adds independent safety verification before responses are released.

The description states that the biggest challenge was designing lightweight safety mechanisms that improve decision quality without introducing unnecessary delays.

There is no evidence of scalability, deployment architecture, or delivery mechanisms beyond a prototype built for a hackathon.

The project is described as a prototype submitted to the OpenAI 2026 hackathon. There is no evidence of:

• Revenue
• Customers
• Adoption
• Product-market fit
• Commercial deployment
• User feedback or usage metrics

It is not evident whether the prototype has been tested in real-world environments or integrated into existing systems.

The description does not mention any competitors or direct market context. It is unclear whether there are existing solutions for AI safety infrastructure, and no evidence of competitive positioning or differentiation is provided.

Inference The project appears to be a novel idea within the AI safety space, but without more information, it's hard to assess its place in the market.

• No traction or commercialization: The product is described as a hackathon prototype with no evidence of revenue, customers, or adoption.
• Unverified claims: All features and functionality are self-reported without independent validation.
• Single founder: The project is built by one person (Arshad Khan), which raises questions about scalability and team capacity.
• Lack of technical detail: The description does not explain how safety mechanisms work, how context is verified, or what the "Human-First" approach entails.
• No pricing or business model: No indication of monetization strategy or commercial viability.

1. What specific AI models are used for context verification and risk assessment?
2. How does the system determine when to involve human oversight?
3. Can you explain how the safety checks are implemented technically?
4. What is the expected latency introduced by these safety mechanisms?
5. Have you tested the prototype in any real-world or simulated environments?
6. What is the roadmap for moving from a prototype to a scalable product?
7. How do you plan to monetize this infrastructure?