How AI Technology Powers FutureFeed
Learn how FutureFeed uses AI to support compliance and document processing, and how your data is protected.
Overview
FutureFeed uses AI to automate key parts of the document workflow — including tagging, classification, and search. This helps you spend less time on manual processing and more time on the compliance work that matters.
All platform AI models are hosted on Amazon Bedrock, a fully managed AWS service that provides secure, scalable access to foundation models without exposing your data to third-party training pipelines.
Current models (through August 27, 2026)
| Capability | Model | Provider |
|---|---|---|
| Document tagging (large documents) | Claude 3.7 Sonnet | Anthropic |
| Document tagging (standard) | Claude 3 Haiku | Anthropic |
| Document classification | Amazon Nova Micro | Amazon |
| Search and retrieval | Amazon Titan Embed Text v2 | Amazon |
Planned models (from August 28, 2026)
| Capability | Model | Provider |
|---|---|---|
| Document tagging (large documents) | To be confirmed | TBC |
| Document tagging (standard) | To be confirmed | TBC |
| Document classification | Amazon Nova Micro | Amazon |
| Search and retrieval | Amazon Titan Embed Text v2 | Amazon |
Note — Anthropic model transition: On February 27, 2026, the U.S. Department of War designated Anthropic as a supply chain risk and directed federal agencies and defense contractors to phase out use of Anthropic products within six months. Because FutureFeed serves the defense supply chain, we are actively evaluating replacement models for the two Anthropic-based capabilities listed above. We will confirm the replacement models and communicate the transition plan to customers well ahead of the August 27, 2026 compliance deadline. This situation is evolving — Anthropic is currently contesting the designation in federal court — and we will update this article as circumstances change.
How We Use AI Across the Business
Product Development
FutureFeed's engineering team uses AI-assisted development tools including Cursor and Anthropic Claude Code to write, review, and improve code more efficiently. These tools operate separately from the platform and do not interact with your data.
General Productivity
FutureFeed team members use ChatGPT (by OpenAI) as a general productivity tool for internal tasks such as drafting communications, brainstorming, and research. ChatGPT is used for internal business purposes only and is not connected to the FutureFeed platform or your data.
Data Privacy and Security
- No model training on your data. Amazon Bedrock does not use your inputs or outputs to train or improve foundation models.
- Data stays in AWS. All AI processing occurs within FutureFeed's AWS Gov Cloud infrastructure. Your documents are not sent to external third-party services.
- Encryption in transit and at rest. All data exchanged with AI models is encrypted using industry-standard protocols.
- Access controls. AI model access is governed by the same role-based permissions that protect the rest of your FutureFeed environment.
Frequently Asked Questions
Can I opt out of AI processing? The AI models are core to FutureFeed's document handling workflow. If you have concerns about specific use cases, contact our support team to discuss your requirements.
Does AI make final decisions on compliance? No. AI assists with tagging, classification, and retrieval to save you time. All compliance decisions remain with you and your team.
Will the AI models change over time? We continuously evaluate and improve the AI models we use. Any changes are tested thoroughly before deployment, and we communicate significant changes through our release notes.