Frequently asked questions
Calai Health provides real-time coaching and a wealth of educational resources, including trainings, workshops, handouts, and informational videos, to support individuals in their recovery journey. The real-time component works by combining contextual understanding with personalized feedback. The system provides immediate, tailored guidance aligned with each person's behavioral and emotional profile. Built on a model fine-tuned with dialectical behavior therapy principles, it continuously adapts through feedback loops and real-time sentiment analysis to ensure trustworthy, clinically-aligned support for each user's unique journey.
Dialectical Behavior Therapy (DBT), is an evidence-based outpatient treatment approach developed by Marsha Linehan in the 1980s for individuals with borderline personality disorder, though it has since been adapted for various other conditions including depression, eating disorders, and substance abuse. DBT combines standard cognitive-behavioral techniques with mindfulness practices and distress tolerance skills, emphasizing the concept of "dialectics" - the ability to hold two seemingly opposing truths simultaneously, such as accepting yourself as you are while also working toward change. The therapy is structured around four core skill modules: mindfulness (staying present and aware), distress tolerance (managing crisis situations without making them worse), emotion regulation (understanding and managing intense emotions), and interpersonal effectiveness (maintaining relationships while getting needs met).
You can sign up for Calai Health by subscribing right now to learn when we go live and then downloading the app when notified.
Yes, your information is protected using industry-leading standards. We’ve implemented SOC 2 and HIPAA-compliant architecture, including end-to-end encryption, secure data storage, and rigorous security audits. All data transmission occurs through encrypted channels, and our infrastructure is protected by multiple layers of security to prevent unauthorized access. We never sell your data and you maintain full control over your information at all times.
Absolutely not. This app does not replace mental health professionals and is not intended to provide clinical diagnosis or individual psychotherapy. Our AI-powered app is designed as an adjunct to professional mental health services and is meant to be used in conjunction with human care from qualified therapists, counselors, or other mental health providers. While our technology offers valuable support, guidance, and resources, it does not substitute for the expertise, clinical judgment, and personalized care that licensed mental health professionals provide. However, it can save you time and money by reducing the need for frequent and costly check-ins with your providers. We strongly encourage users to maintain a regular schedule with their healthcare providers and to seek immediate help for emergencies or crises.
Our AI-native mental health platform is built on a foundation of clinical expertise, with a licensed clinical psychologist serving as the lead Chief Clinical Officer (CCO) working arm-in arm with our Chief Technical Officer (CTO) as the co-architects for our machine learning systems. The CCO doesn't just review the final product—they're embedded throughout the development process, defining the training parameters, curating datasets, and establishing guardrails that ensure every interaction aligns with evidence-based therapeutic principles.
Technically, this means our AI models are trained with carefully filtered data that excludes harmful content patterns, while positive therapeutic frameworks are reinforced through supervised learning techniques. The CCO and CTO have programmed specific safety protocols that actively detect and prevent discussions that could promote disordered eating, body shaming, anti-fat bias, or weight-focused interventions. Instead, the system is trained to recognize these topics and redirect conversations toward body-neutral language and healthy coping mechanisms.
The AI learns not just what to say, but what never to say. Through a process called constrained optimization, our models are designed to maximize therapeutic benefit while operating within strict ethical boundaries. This creates an AI that can provide nuanced mental health support while maintaining the same duty of care and "do no harm" principles that guide human therapists. The result is technology that is intuitive and supportive, backed by the rigor and wisdom of clinical psychology practice.
This app was conceived by a licensed clinical psychologist and eating disorders expert with decades of professional experience. The lead technical implementation was conducted by a machine learning architect with a strong track record building and scaling HIPAA-compliant, SOC 2-ready platforms. With deep expertise in LLM-embedded systems and AI-native products, the engineering lead ensured that the app’s architecture is robust, intelligent, and designed to evolve with user needs. This collaboration between comprehensive clinical expertise and advanced AI technological competency ensures that our app delivers both evidence-based support and cutting-edge functionality for users.
AI-native applications like Calai possess a fundamentally different scalability profile compared to traditional software. As these applications process more data, they don't just handle increased volume—they actually become more intelligent and capable. The underlying machine learning models continuously learn from new inputs, refining their accuracy, expanding their knowledge base, and developing more nuanced understanding of user patterns and preferences.
This creates a compounding advantage where scale drives quality improvements rather than just operational efficiency. More users generate more diverse data points, which enable the AI to handle edge cases better, reduce errors, and discover new patterns that enhance functionality for all users. Unlike conventional applications that may slow down or become more complex with scale, AI-native apps become more responsive and personalized as they accumulate training data.
The network effects are particularly powerful—each additional user interaction contributes to a shared intelligence that benefits the entire user base. This means AI-native applications can achieve both horizontal scaling (serving more users) and vertical scaling (becoming smarter and more valuable) simultaneously, creating sustainable competitive moats that strengthen over time rather than requiring constant feature development to maintain relevance.
Our ultimate goal is to dramatically increase access to quality care by making it both affordable and instantly available, ensuring that cost and availability barriers don't prevent anyone from getting the support they need during their most vulnerable moments. We’re launching as a B2C-first company, targeting individuals who need fast, evidence-based support, while navigating eating disorder recovery. This strategy allows us to validate product-market fit in a high-impact, urgent-use environment. From day one, the platform has been built with enterprise scalability in mind. Once we've proven traction and outcomes in the consumer space, we’ll expand into B2B opportunities, including licensing partnerships with treatment centers, healthcare providers, employers, and payers. Our long-term goal is to become a category-defining solution trusted by patients and institutions - driven by data, backed by outcomes, and ready to scale across sectors.