How is CORTEX different from a cloud chatbot?
Cloud assistants are a chat box that lives in their browser. CORTEX is an operating layer that lives on your Mac, iPhone, and Watch, with a permanent memory you own, an audit trail of every action, and a real shell that touches your filesystem. The frontier models underneath are interchangeable plumbing. CORTEX is the system that uses them.
Do you use my data to train any model?
No. Never. Your conversations, files, calendar, finances, and audit trail are stored in your local memory and your private cloud region. They are not sent to any vendor for training. The frontier models we route to operate under zero-retention API agreements where available.
Is there really a free tier? What's the catch?
Yes. The Personal tier is free forever. It runs a local model on your Mac, with the full execution layer, memory, audit trail, and autonomous loop. No cloud calls. No subscription. No catch. Pro tier exists because frontier cloud reasoning costs real money. We absorb the vendor costs so you pay one flat fee.
What happens to my data if I cancel?
Your data is yours. On cancel, you can export your full memory, audit trail, and connector history as portable JSON. Your local Personal-tier brain keeps running. Your iCloud-synced files stay in iCloud. We hold nothing hostage.
Can CORTEX run without an internet connection?
The Personal tier runs fully offline. Local model, local memory, local execution. Pro, Founder, and Apex tiers degrade gracefully: when the cloud is unreachable, CORTEX falls back to the local model and queues cloud-bound actions for replay when connectivity returns.
Which models does CORTEX use under the hood?
CORTEX is provider-agnostic. Cloud-class frontier reasoning powers the Pro tier, quality-lane reasoning powers the Founder tier, a local model runs offline. Vision, OCR, image generation, transcription, and voice synthesis all route through best-in-class providers selected by capability and cost. You don't see the routing. You see one CORTEX.
How do I switch tiers? Annual vs monthly?
Switch any time from Settings. Annual billing saves ~20% vs monthly on every paid tier. Your memory, agents, and connectors carry across. Downgrade returns you to fair-use caps; nothing is deleted.
Is CORTEX HIPAA-compliant? SOC 2?
CORTEX architecture aligns with SOC 2 Type 2, GDPR, CCPA, and HIPAA principles. AES-256-GCM at rest, TLS-pinned in transit, hash-chained audit, role-based access, no training on your data. Formal SOC 2 audit is underway in 2026. HIPAA-ready offering is available on the Apex tier with a Business Associate Agreement.
How is CORTEX priced compared to running these models myself?
If you wired up the underlying frontier models, vision, image generation, transcription, voice synthesis, and cloud infrastructure yourself, you'd be juggling six accounts and paying à la carte for every call. CORTEX absorbs the vendor cost, the iOS/Mac/Watch surface, the security stack, and the support, and gives you one flat number per tier. You pay for the system, not the API calls.
Do you offer education or nonprofit pricing?
Yes. 50% off Pro for verified students, educators, and registered 501(c)(3) nonprofits. Email
hello@cortexnode.ai with verification, and a human (usually George) approves it that day.
Who built CORTEX? Why should I trust this?
CORTEX is built by George Bayze, a single founder running a real business in Tucson, Arizona. Every line of architecture, every governance law, every commit. Owner-operator. CORTEXNODE Inc. is the parent company. The product was reviewed and validated by senior leadership at a major U.S. consumer-tech company in April 2026. The full operating doctrine is public.
