CRITICAL DISCLAIMER: Perfect security is mathematically and practically impossible. Our threat model has explicit limitations:
ATTACKS WE CANNOT PREVENT:
Endpoint compromise: Malware on user device (keyloggers, clipboard hijacking, memory dumps, screen capture). If your device is compromised before encryption, no cloud service can help.
Physical coercion: Rubber-hose cryptanalysis, $5 wrench attacks, torture, detention. We cannot protect against threats to your physical safety.
Advanced nation-state attacks: Supply chain attacks on hardware/firmware, CPU vulnerabilities (Spectre, Meltdown variants), hardware implants, TEMPEST attacks, timing analysis.
Social engineering: Phishing, pretexting, baiting. Human factors remain the weakest link.
Quantum computing (future): While we implement post-quantum cryptography (Kyber-768), the field is evolving. Harvest-now-decrypt-later attacks could threaten today's encrypted data decades from now.
Zero-day exploits: Unknown vulnerabilities in our codebase, dependencies, or cryptographic implementations. Open source helps, but doesn't eliminate this risk.
Traffic analysis: Despite Tor/VPN usage, sophisticated adversaries can potentially correlate upload/download patterns, file sizes, timing.
Infrastructure attacks: DDoS, BGP hijacking, DNS poisoning, submarine cable tapping. We mitigate but cannot fully eliminate.
OUR DEFENSE STRATEGY:
Multi-entity architecture: Backend fragmented across 5+ legal entities in different jurisdictions. Each entity operates independent infrastructure, performs code verification, and adheres to immutable security principles encoded in founding documents. No single entity has complete access.
Mutual verification: Each backend entity must cryptographically sign and verify others' code deployments. Rogue updates are detectable. Consensus required for infrastructure changes.
Jurisdictional isolation: Frontend, API, storage, payment processing are legally and physically separate. Compromise of one component doesn't cascade.
Immutable security policies: Core principles (zero-knowledge, no logging, encryption standards) written into legal structure. Cannot be changed without dissolving and reconstructing entities - economically infeasible.
Cryptographic agility: Hybrid classical/post-quantum scheme allows algorithm updates. Monitoring NIST post-quantum standards for migration triggers.
Active threat intelligence: Continuous monitoring of: CVE databases, cryptographic research, jurisdiction legal changes, nation-state capabilities. Incident response plans for various attack scenarios.
FUTURE ADAPTATION COMMITMENT:
Security is not static. We commit to: quarterly security audits, immediate response to critical vulnerabilities, transparent disclosure of incidents affecting users (within legal constraints), regular cryptographic algorithm updates, expansion to new jurisdictions as needed, hardware security module integration (planned), threshold cryptography for enterprise (researching).
USE CASE APPROPRIATENESS:
tesho is suitable for: activists in hostile environments, journalists protecting sources, businesses with sensitive IP, individuals seeking privacy from corporate surveillance, data requiring multi-decade confidentiality.
tesho is NOT suitable for: protection against targeted nation-state attacks on high-value individuals (use airgapped systems), scenarios where physical security is compromised, situations requiring absolute certainty (nothing provides this).
TRANSPARENCY: This disclosure exists because we respect your intelligence. Services claiming "unhackable" or "military-grade" without caveats are either incompetent or dishonest. Security requires informed users making risk-aware decisions.