Ideas catalog
Every idea we've derived from recent research — the ones marked “Prototype” come with a live demo you can try.
A niche hardware play targeting fleet emissions compliance, though it faces significant mechanical complexity and a potential conflict between emission reduction and tire longevity.
Backed by 2 papers
A promising technical approach to edge AI, but currently framed as a generic service rather than a specific product, making it a feature for chipmakers rather than a standalone business.
Backed by 2 papers
A niche B2B tool with clear utility for compliance, but currently lacks a concrete predictive mechanism beyond a general literature review.
A high-risk clinical service model that relies on a substance currently lacking broad regulatory approval and faces significant operational hurdles despite promising efficacy data.
Backed by 3 papers
A high-potential clinical tool that is currently undermined by a critical mismatch between the theoretical mathematical model provided and the practical biological data required for implementation.
Mechanistically sharp companion diagnostic concept riding the GLP-1RA wave, but the commercial thesis rests almost entirely on mouse knockout data with only indirect human metabolite evidence, leaving a large clinical-validation gap before any CDx submission is conceivable.
Backed by 2 papers
A high-friction hardware retrofit that addresses a looming regulatory cliff (EURO 7), though it faces significant engineering hurdles regarding vehicle aerodynamics and maintenance.
Backed by 3 papers
A highly specialized diagnostic tool with strong scientific grounding but severe practical bottlenecks due to the cost and accessibility of longitudinal fMRI.
A focused, clinically validated AI plug-in for standardizing midpalatal suture maturation scoring—a real diagnostic pain point—but it solves a narrow sub-problem that fits better as a feature inside larger imaging platforms than as a standalone venture.
Backed by 2 papers
A niche but viable optimization for automotive HVAC that leverages federated learning to balance personalization with privacy, though it faces high integration hurdles.
A niche HR training tool that translates a small-scale academic study into a corporate compliance product, though it faces significant headwinds regarding buyer urgency and scalability.
A generic digital health application that translates a well-known clinical protocol into a standard app, lacking a unique technical moat or a specific high-urgency wedge.
A niche utility tool that leverages a specific mission's data, but lacks a clear commercial business model given that the target beneficiaries typically build these tools in-house.
A useful but generic application of existing IPCC data that lacks a proprietary technical moat and faces high friction in municipal procurement.
A high-risk psychological 'trick' that leverages a single study to deceive users into thinking an AI is human, creating significant ethical and regulatory liabilities in the mental health space.
A clinically grounded orthodontic outcomes registry with a real niche, but it faces hard data-aggregation hurdles and an uncertain willingness-to-pay in a relatively narrow vertical.
Backed by 3 papers
A moderately useful UI enhancement for BI tools that lacks a strong competitive moat and faces significant headwinds from existing 'drill-down' features in market leaders.
Compelling clinical evidence and a clear unmet need are fatally undermined by the FDA's August 2024 Complete Response Letter rejecting Lykos's NDA, which renders the 'why now' premise commercially inert for the foreseeable future.
A technically sound, well-targeted bean-breeding program that smartly exploits a clean monogenic resistance trait, but it relies on a single study, uses standard (non-novel) breeding techniques, and faces a low-margin, donor-dependent smallholder seed market.
A low-moat consultancy play that repackages existing public IPCC data into reports, lacking a scalable product edge or proprietary technology.