ChatLibris

What the company appears to be: ChatLibris is a self-reported academic research assistant that searches indexed academic literature, synthesizes retrieved papers, and returns either a cited answer or an explicit "Unknown" when evidence is insufficient. It was built as a hackathon project by one person (DrHCodes Milojevic) with no user accounts, database, conversation history, or full-text parsing capabilities.

What changed: The author states they focused on building a system that prevents fabricated citations and enforces abstention from answering when evidence is lacking — distinguishing between technical failures and scientific uncertainty. This approach was designed to be "trustworthy AI" by building trust into the architecture rather than adding disclaimers afterward.

The single most important open question: Does ChatLibris have any commercial traction, revenue, or customer adoption beyond its hackathon prototype? The description makes no claims about monetization, users, or market validation — only that it was built under time constraints and is intended to evolve toward more advanced features.

The description states that ChatLibris:

• Searches an academic index
• Synthesizes only the retrieved papers
• Says "unknown" when evidence cannot support an answer
• Prevents fabricated citations by generating source identifiers itself and checking model outputs
• Distinguishes between different types of failure (e.g., no sufficient literature, synthesis failure)
• Restricts retrieval to indexed academic literature (not necessarily peer-reviewed)
• Was built for a hackathon with limited scope

It is described as a system that "prevents the system from answering when the evidence was insufficient" and enforces structured output.

Inference: Based on the author's own account, this is not a production-ready product but a proof-of-concept prototype. It lacks user accounts, persistent storage, conversation history, or PDF upload capabilities.

The description states:

• ChatLibris is positioned as an assistant that "knows when it should not answer"
• It distinguishes between technical failure and scientific uncertainty
• The goal is to make academic evidence easier to access without stripping away uncertainty, limitations, or disagreement
• It was built with the understanding that “trustworthy AI cannot be created by adding a disclaimer after generation”
• Future versions may include stronger peer-review filtering, systematic-review prioritization, study-design classification, and full-text analysis

Inference: The positioning evolved from a hackathon prototype to a vision of a trustworthy research assistant that respects scientific uncertainty. However, the current version is not described as having any commercial or user-facing features beyond basic question-answering.

The description does not state:

• Who the target customer is
• Whether there is an identified ICP (Ideal Customer Profile)
• If there are any existing users or personas defined

Not evidenced: No information about who uses ChatLibris, what their needs are, or how they would interact with it beyond the author’s own use case.

The description does not state:

• How ChatLibris intends to make money
• Whether there is a pricing model
• If any monetization strategy has been developed or tested

Not evidenced: No evidence of business model, pricing, or revenue streams. The project is described as a hackathon submission with no indication of commercial intent.

The description states:

• Built using OpenAI Codex and Vercel
• Uses server-side rules to enforce abstention
• Generates source identifiers internally to prevent fake citations
• Distinguishes between multiple failure modes (search, synthesis, timeout)
• Was built under time pressure with aggressive scope cuts
• Did not include user accounts, databases, conversation history, PDF uploads, vector databases, or full-text parsing

Inference: The technical architecture is minimal and hackathon-focused. It does not appear to be production-ready or scalable.

The description states:

• This was a hackathon project (OpenAI 2026)
• No user accounts, database, conversation history, or persistent storage
• No revenue, customers, or adoption data
• The author explicitly says they cut scope aggressively for the hackathon

Not evidenced: No evidence of traction, headcount, customer base, or usage metrics. The project is described as a prototype with no commercial deployment.

The description does not state:

• Who the competitors are
• What existing products or services exist in this space
• How ChatLibris differentiates from them

Not evidenced: No competitive analysis or market positioning beyond self-reported claims about trust and uncertainty.

Key risks and open questions based on the description:

• The project is a hackathon prototype with no commercial traction, revenue, or users
• It lacks core features like user accounts, persistent storage, conversation history, or full-text analysis
• There is no evidence of any business model or monetization strategy
• The author states that most of the intended functionality was not built due to time constraints
• No third-party validation or independent verification of claims

Inference: The project appears to be a proof-of-concept with no clear path to market traction or commercial viability.

1. What is your plan for monetization and customer acquisition beyond the hackathon?
2. Have you validated demand from potential users in academic or research settings?
3. How do you intend to scale beyond a single-user prototype?
4. What are the technical limitations of the current architecture that would prevent production use?
5. Are there any partnerships or institutional relationships that support this project’s future development?