Built for law firms and legal teams that cannot afford mistakes. Gideon is in active alpha, with the current focus on secure foundations, public transparency, and a controlled path to early access.
Already invited? Login to the app
Why teams start here
Gideon is aimed at legal teams that want AI help without giving up confidentiality, auditability, or clear product boundaries between the public website and the authenticated app.
Status: Active Alpha Development. Gideon is being built in public, with security, compliance, and transparency prioritized before broad access.
Q2 2026 focus
Q3 2026 and beyond
Interested in Gideon's direction? Broad access is not open yet, but you can follow development on GitHub.
Multi-tenant cryptographic isolation from day one. Designed for professionals handling attorney-client privileged information. Every interaction is logged for compliance and audit trails.
Designed to meet SOC 2 Type II requirements. Role-based access control, data encryption at rest and in transit, audit logs with UTC timestamps. Certification planned for Q4 2026.
Public GitHub repository. All architectural decisions documented (ADRs). Community-driven roadmap. No black boxes. See exactly how it works.
AWS Bedrock provides access to multiple LLM providers (Claude, GPT, etc.). Modular architecture allows easy model switching. Export your data anytime.
Legal work is high stakes. Deadlines, confidentiality, and professional responsibility leave very little room for vague answers or brittle tools. Gideon is being built to give legal professionals useful AI assistance without asking them to trade away control, auditability, or trust.
That means starting with secure foundations, transparent product status, and open development instead of polished promises. Gideon is still in active alpha, but the direction is clear: practical AI for professionals who cannot afford mistakes.
The next useful step is not self-serve signup yet. It is a clear handoff for early-access requests, design-partner conversations, and existing invited-user login.
Gideon is not ready for broad self-serve access yet. If you want to be considered for alpha access, start with the public early-access handoff and a direct conversation.
What this means today:
The codebase is public. Follow the roadmap, inspect the documentation, and track how the product evolves before broader access opens.
Useful next steps:
If you are evaluating Gideon for a future pilot, design-partner role, or strategic introduction, talk directly before any account or billing flow exists.
We're open to:
A clear view of what is live today, what is actively being built, and what comes next.
AWS Bedrock integration with Claude, real-time responses, context-aware conversations
AWS Cognito user pools, multi-factor authentication support, session management
Cryptographic data separation, tenant-specific workspaces, privilege protection
Immutable activity logs, UTC timestamp precision, CloudWatch integration
MCP-based plugin system, custom tool integration, API-first design
Save conversations across sessions, search previous interactions, sidebar navigation
Token-by-token response display, interrupt long responses, progress indicators
Rate AI responses, flag inaccurate information, continuous improvement loop
Analyze legal documents, extract key clauses, summarize findings
Case law search (jurisdiction-aware), statute lookup, regulatory guidance
Clause identification, risk assessment, comparison tools
Automatic source attribution, Bluebook format support, verification links
Connect to practice management software, matter-based context, team collaboration
The roadmap stays public so legal teams and contributors can see what is shipping, what is being hardened, and what is still ahead.
⚠️ Timeline Disclaimer: Dates are directional and may change based on product learning, technical constraints, and contributor capacity. GitHub reflects the freshest progress signal.
Enterprise-grade infrastructure designed for security, scalability, and compliance.
Founder & Builder
Building Gideon for legal professionals who need AI systems they can trust under pressure, with security, compliance, and transparency treated as product fundamentals.