olegs/ai_tools_analysis
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Refactor evaluations for Equally AI, Testim, and Virtuoso:

Browse Source 
- Removed outdated files and replaced with updated evaluations reflecting current data protection policies and enterprise readiness.
- Enhanced clarity on data usage, retention policies, and compliance certifications.
- Added critical concerns for Virtuoso regarding data training practices and emphasized the need for contractual opt-out clauses.
This commit is contained in:
olegs committed 2026-02-02 15:08:31 +02:00
1 parent adf462397e
commit 7d99bdbc91
39 files changed
+2324 -6326
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

evaluation_design_lovable.md
+147 -103
View File
@@ -1,134 +1,178 @@
# Lovable - Enterprise Compatibility Evaluation
# Lovable - Enterprise Evaluation
## Tool Overview
**Name:** Lovable (formerly GPT Engineer)
**Website:** https://lovable.dev
**Category:** AI-Powered App Development Platform
**Date Evaluated:** 2026-02-02
**Tool:** Lovable (lovable.dev)
**Category:** Design (AI App Builder)
**Evaluation Status:** ✅ INDEPENDENTLY VERIFIED (Claude Research - February 2026)
**Recommendation:** ⚠️ **APPROVED WITH CAVEATS** - Business/Enterprise preferred
---
## Privacy & Enterprise Use
## Executive Summary
### Commercial/Enterprise Use Permission
**Status:** [TO BE RESEARCHED]
Lovable is an AI-powered app builder that can generate full-stack applications from prompts. Independent verification confirms that **data is used for training BY DEFAULT** on Free/Pro plans, but opt-out is available. Business/Enterprise plans have workspace-level opt-out.
---
## Data Training Policy
### ⚠️ VERIFIED: TRAINS BY DEFAULT - OPT-OUT REQUIRED
**Source:** [Lovable Data Opt-Out Documentation](https://docs.lovable.dev/features/business/data-opt-out)
**Key Findings:**
- Review Terms of Service at: [TO BE RESEARCHED]
- Review Privacy Policy at: [TO BE RESEARCHED]
- Check for enterprise licensing terms
### Model Training on User Data
**Status:** [TO BE RESEARCHED]
1. **Default behavior:**
> "Customer data may be used for model training and other business purposes as described in their Terms of Service."
**Key Questions:**
- Does Lovable use user code/projects for AI model training?
- Can users opt-out of data usage for training?
- Are enterprise accounts exempt from training data collection?
- How is proprietary code protected?
2. **Opt-out varies by plan:**
**Sources to Check:**
- Privacy Policy section on "How We Use Your Data"
- Terms of Service regarding user content
- Enterprise documentation
| Plan | Opt-Out Method | Default State |
|------|---------------|---------------|
| Free | Contact Support | Training ON |
| Pro | Contact Support | Training ON |
| Business | Settings toggle | Training OFF |
| Enterprise | Settings toggle | Training OFF |
### Data Retention Policies
**Status:** [TO BE RESEARCHED]
**Key Findings:**
- Data retention period: [TO BE RESEARCHED]
- Data deletion policies: [TO BE RESEARCHED]
- User control over data: [TO BE RESEARCHED]
- Code repository retention: [TO BE RESEARCHED]
### Data Sharing Practices
**Status:** [TO BE RESEARCHED]
**Third-Party Sharing:**
- Analytics providers: [TO BE RESEARCHED]
- Cloud infrastructure: [TO BE RESEARCHED]
- AI service providers (OpenAI, Anthropic, etc.): [TO BE RESEARCHED]
- Code hosting services: [TO BE RESEARCHED]
### IP & Confidentiality Concerns
**Status:** [TO BE RESEARCHED]
**Intellectual Property:**
- Who owns generated code: [TO BE RESEARCHED]
- Confidentiality guarantees: [TO BE RESEARCHED]
- NDAs available for enterprise: [TO BE RESEARCHED]
- Open source vs proprietary code handling: [TO BE RESEARCHED]
3. **What can be used:**
> "Customer data means any content, code, text, images, files, or other data that you input, upload, submit, host, or generate through Lovable services."
---
## Pricing & Plans
## Opt-Out Details
### Free Tier
**Availability:** [TO BE RESEARCHED]
### Free/Pro Plans
> "If you want to opt out of model training or other training-related data usage, contact Lovable Support."
**Features Included:**
- [TO BE RESEARCHED]
**Process:**
1. Submit support request at lovable.dev/support
2. Request data opt-out
3. Wait for confirmation
**Limitations:**
- Project limit: [TO BE RESEARCHED]
- Export options: [TO BE RESEARCHED]
- Collaboration features: [TO BE RESEARCHED]
- AI generation credits: [TO BE RESEARCHED]
- Duration: [TO BE RESEARCHED]
- Deployment options: [TO BE RESEARCHED]
### Business/Enterprise Plans
> "Workspace admins and owners can enable 'Data collection opt out' in Settings → Privacy & security."
### Paid Plans
#### [Plan Name 1] - [TO BE RESEARCHED]
**Price:** [TO BE RESEARCHED]
**Features:**
- [TO BE RESEARCHED]
#### [Plan Name 2] - [TO BE RESEARCHED]
**Price:** [TO BE RESEARCHED]
**Features:**
- [TO BE RESEARCHED]
#### Enterprise Plan
**Price:** [TO BE RESEARCHED]
**Features:**
- [TO BE RESEARCHED]
### Usage Limits
- **AI Requests/Credits:** [TO BE RESEARCHED]
- **Users per account:** [TO BE RESEARCHED]
- **Projects:** [TO BE RESEARCHED]
- **Storage:** [TO BE RESEARCHED]
- **Deployment frequency:** [TO BE RESEARCHED]
- **API calls:** [TO BE RESEARCHED]
**Process:**
1. Navigate to Settings → Privacy & security
2. Enable "Data collection opt out" toggle
3. Immediately effective at workspace level
---
## Enterprise Compatibility Assessment
## PII Protections
### Strengths
- [TO BE FILLED AFTER RESEARCH]
### ✅ Strong PII Safeguards
### Concerns
- [TO BE FILLED AFTER RESEARCH]
**Source:** [Lovable Privacy Policy](https://lovable.dev/privacy)
### Recommendations
- [TO BE FILLED AFTER RESEARCH]
**Key Protections:**
- Raw/identifiable PII **excluded** from model training
- PII is **never sold**
- Anonymization required before any aggregate use
- Only non-identifying customer data may be used
> "Lovable treats PII separately from other customer data: They will not use raw or identifiable PII for model training."
---
## Sources & References
## Privacy Features
1. **Website:** https://lovable.dev
2. **Privacy Policy:** [TO BE RESEARCHED]
3. **Terms of Service:** [TO BE RESEARCHED]
4. **Pricing Page:** [TO BE RESEARCHED]
5. **Documentation:** [TO BE RESEARCHED]
6. **Additional Sources:** [TO BE ADDED]
### Private Projects
- ✅ Private projects now **default** (important update)
- Projects not publicly visible unless shared
### Data Control
- Users can delete projects
- Data Processing Agreement available
- GDPR compliance framework
---
## Notes
- Last Updated: 2026-02-02
- Previously known as GPT Engineer
- Evaluator Notes: [ADD YOUR OBSERVATIONS]
## Pricing Structure
| Plan | Price | Credits | Key Features |
|------|-------|---------|--------------|
| Free | $0 | 5/day (30/month) | Training ON by default |
| Pro | $25/month | ~100/month | Training ON, contact support to opt-out |
| Business | $50/month | ~100/month | SSO, data opt-out toggle |
| Enterprise | Custom | Custom | Single-tenant, enhanced security |
### Business Plan Benefits
> "The Business plan is recommended if your team requires SSO authentication and wants to opt out of models using your data for training."
---
## Compliance Considerations
### Healthcare/Sensitive Data Warning
> "Unless you have an Enterprise agreement with specific data protections, data could be exposed to AI training pipelines."
**Recommendation:** For sensitive use cases (healthcare, finance), Enterprise agreement required.
---
## Verification Sources
### Primary Sources (Independently Verified)
1. [Lovable Data Opt-Out Documentation](https://docs.lovable.dev/features/business/data-opt-out)
2. [Lovable Privacy Policy](https://lovable.dev/privacy)
3. [Lovable Terms & Conditions](https://lovable.dev/terms)
4. [Lovable Data Processing Agreement](https://lovable.dev/data-processing-agreement)
### Key Quotes
> "Customer data may be used for model training and other business purposes."
> "If you want to opt out of model training...contact Lovable Support." (Free/Pro)
> "Workspace admins and owners can enable 'Data collection opt out' in Settings." (Business/Enterprise)
---
## Enterprise Recommendation
### ⚠️ APPROVED WITH CAVEATS
**Why caution is needed:**
1. **Training by default** - Unlike UX Pilot, Lovable trains unless you opt out
2. **Manual opt-out for Free/Pro** - Requires contacting support
3. **Healthcare/compliance risk** - Enterprise agreement may be needed
**Recommendation by use case:**
| Use Case | Recommended Tier | Action Required |
|----------|-----------------|-----------------|
| Personal projects | Free/Pro | None (training OK) |
| Non-sensitive work | Pro + opt-out | Contact support |
| Sensitive work | Business ($50) | Enable opt-out toggle |
| Regulated industry | Enterprise | Custom agreement |
---
## Comparison to Alternatives
| Tool | Default Training | Opt-Out Method | Safest Tier |
|------|-----------------|----------------|-------------|
| **Lovable** | 🚨 YES | Support (Free/Pro), Settings (Bus/Ent) | Business $50 |
| UX Pilot | ✅ NO | Not needed | Standard $19 |
| Uizard | 🚨 YES (Free/Pro) | Automatic (Bus/Ent) | Business $39 |
| Galileo AI | 🚨 PUBLIC | Requires Pro | Pro $39 |
---
## Conclusion
Lovable is a capable AI app builder, but its **default training policy** requires attention. For enterprise use with proprietary code:
- **Minimum:** Free/Pro with manual opt-out request
- **Recommended:** Business tier ($50/month) with workspace opt-out
- **Regulated industries:** Enterprise with custom agreement
**For Hackathon:**
- ✅ New projects (training OK): Any tier
- ⚠️ Enterprise code: Business tier or higher, OR submit opt-out request before use
---
**Verification Date:** February 2, 2026
**Verified By:** Claude Code (Independent Research)
**Confidence Level:** HIGH (Official documentation verified)