Client Data Promise: We do not use your proprietary data to train public LLM models. Your data is isolated and only used for delivering our Services to you. Individual Customer Data is never shared between customers or used to train third-party AI models without explicit consent.
Gneis Agency ("Gneis Agency", "we") is committed to transparent and responsible use of artificial intelligence (AI) in our SaaS platform and services (the "Services"). This policy describes how AI is used in the Services, what data is processed, and what rights and responsibilities customers and users have.
This policy supplements our Terms and Conditions, Privacy Policy and Data Processing Agreement.
1. AI as Core Component
1.1 AI-Driven Platform
Gneis Agency's Services are fundamentally based on AI technology. The platform uses large language models (LLMs) and machine learning to:
- Analyze and score visibility across AI-driven search and answer systems
- Generate insights, recommendations and diagnostics
- Benchmark performance against competitors
- Deliver predictive analytics and trends
- Generate playbooks and action recommendations
1.2 Third-Party Providers
Gneis Agency uses advanced AI models from leading third-party providers:
- OpenAI (GPT series)
- Anthropic (Claude series)
- Google (Gemini/PaLM)
We select models based on:
- Quality and accuracy
- Compliance and security
- Performance and speed
- Specialization for specific tasks
2. How AI is Used in the Services
2.1 Input Processing (Customer Content)
AI models receive and process:
- Prompts and queries: Your keywords, brands, products and markets
- Configurations: Settings, preferences and parameters
- Feeds and data: Structured data, product catalogs, URLs
- Context: History, previous interactions (in session scope)
2.2 Output Generation
AI models generate:
- Scores and ratings: Visibility scores, performance metrics
- Analyses: Detailed reports on strengths and weaknesses
- Recommendations: Concrete improvement suggestions and playbooks
- Benchmarks: Comparison with competitors and industry standards
- Predictions: Trends and future developments
2.3 Continuous Learning and Improvement
Gneis Agency uses aggregated, anonymized data to:
- Improve scoring algorithms and models
- Train and fine-tune internal models
- Validate and calibrate outputs
- Identify new patterns and trends
Important: Individual Customer Data is not shared between customers or used to train third-party LLMs without consent.
3. Data Processing and Privacy
3.1 What is Sent to AI Models?
The following data may be sent to third-party AI providers:
- Prompts, queries and configurations
- Brands, product names and market information
- Publicly available content (URLs, metadata)
- Aggregated, anonymized benchmark data
3.2 What is NOT Sent to AI Models?
The following data is not sent to third-party LLMs:
- Personally identifiable information (PII) without necessity
- Payment information
- Passwords or authentication tokens
- Confidential business data marked as confidential
- Other customers' data
3.3 Data Retention by Third-Party Providers
Our agreements with AI providers specify:
- Zero retention for most requests (data deleted immediately after processing)
- 30-day retention for abuse monitoring with certain providers
- No training on Customer Data without explicit consent
- Compliance with GDPR, EU AI Act and other relevant regulations
3.4 Geographic Processing
AI processing may occur in:
- EU regions (preferred for EU customers)
- USA (when using certain LLM models)
- Other regions with appropriate safeguards
International transfers are secured through EU Commission Standard Contractual Clauses (SCC) and supplementary technical measures.
4. EU AI Act Compliance
4.1 Risk Classification
Gneis Agency's Services are classified as:
- General Purpose AI (GPAI): We use general purpose LLMs from third parties
- Limited Risk AI: The Services generate commercial analysis and recommendations but do not make binding decisions about individuals
Gneis Agency does not provide:
- High-risk AI systems as defined in EU AI Act Art. 6
- Biometric identification or categorization
- AI for law enforcement or migration
- AI for social scoring
4.2 Transparency Obligations
In accordance with the EU AI Act, Gneis Agency commits to:
Transparency:
- Clearly inform that AI is used in the Services
- Describe purposes and limitations of AI outputs
- Document AI models and data sources
- Provide access to this AI Transparency Policy
Quality and Accuracy:
- Validate and calibrate AI outputs regularly
- Implement quality assurance and human review where relevant
- Document error rates and limitations
- Continuously improve models and algorithms
Human Oversight:
- Ensure customers can override or adjust AI outputs
- Provide mechanisms for feedback and correction
- Implement human-in-the-loop where critical
Security and Robustness:
- Protect against adversarial attacks and jailbreaking
- Implement guardrails and content filters
- Monitor for bias, discrimination and unfairness
- Test for edge cases and unexpected inputs
4.3 Customer's Responsibility under AI Act
The Customer is responsible for:
- Assessing whether their use of Output constitutes high-risk AI
- Ensuring compliance with transparency and information obligations to end users
- Implementing appropriate oversight and governance
- Documenting AI use in accordance with AI Act
Gneis Agency is not responsible for Customer's use of AI outputs for high-risk purposes without appropriate safeguards.
5. Limitations and Responsible Use
5.1 AI is Probabilistic
AI-generated outputs are probabilistic and may be:
- Incomplete or inaccurate
- Affected by bias in training data
- Sensitive to prompt formulation
- Non-reproducible (different outputs for same input)
5.2 No Guarantee of Correctness
Gneis Agency does not guarantee:
- That AI outputs are correct, complete or up-to-date
- That recommendations will lead to specific business results
- That scores and benchmarks are absolute or immutable
5.3 Human Responsibility
The Customer is responsible for:
- Reviewing and validating AI outputs before use
- Making business decisions with professional judgment
- Not blindly trusting AI recommendations without critical assessment
- Ensuring compliance with relevant legislation in their use of outputs
5.4 Not Professional Advice
AI outputs do not constitute:
- Legal advice
- Financial advice
- Medical or health advice
- Other professional advice requiring licensing
6. Bias, Fairness and Ethics
6.1 Bias Mitigation
Gneis Agency actively works to:
- Identify and reduce bias in AI models
- Validate outputs across different markets and industries
- Diversify training data and benchmarks
- Implement fairness checks and audits
6.2 Known Limitations
AI models may have:
- Linguistic bias: Better performance in English than other languages
- Geographic bias: Overrepresentation of Western/American data
- Temporal bias: Limited knowledge of very recent developments
- Domain bias: Varying performance across industries
6.3 Ethical Principles
Gneis Agency's AI use is governed by:
- Transparency: Openness about AI use
- Accountability: Clear ownership of outputs and consequences
- Fairness: Striving for equal and impartial treatment
- Privacy: Protection of personal and confidential data
- Safety: Prevention of harm and abuse
- Human oversight: AI as tool, not replacement for human judgment
7. Security and Abuse Prevention
7.1 Security Measures
Gneis Agency implements:
- Input validation: Scanning for malicious prompts and injection attacks
- Output filtering: Blocking harmful, illegal or unethical content
- Rate limiting: Prevention of abuse and overload
- Monitoring: Real-time monitoring for anomalies and attacks
- Incident response: Procedures for security incidents
7.2 Prohibited AI Use
The following is strictly prohibited (see also AUP):
- Jailbreaking or circumventing security mechanisms
- Prompt injection or adversarial attacks
- Reverse engineering of models
- Generation of harmful, illegal or offensive content
- Automated spam or manipulation
- Training competing models
7.3 Third-Party Policies
The Customer must comply with AI providers' policies:
- OpenAI Usage Policies
- Anthropic Acceptable Use Policy
- Google Cloud AI/ML Terms
8. Rights to AI Outputs
8.1 Customer's Rights
The Customer has the right to:
- Use AI outputs for internal business purposes
- Incorporate outputs into own documents and processes
- Share outputs within the organization
- Make business decisions based on outputs
8.2 Limitations
The Customer must not:
- Resell raw outputs as standalone product
- Present outputs as exclusively human-generated without AI involvement (where transparency obligation applies)
- Use outputs to train competing AI models
- Use outputs in ways that violate AUP or third-party policies
8.3 Gneis Agency's Rights
Gneis Agency retains:
- Ownership of underlying models, algorithms and methodologies
- Right to use aggregated, anonymized data for improvements
- Right to generate similar outputs for other customers
- Intellectual property rights to the platform and software
9. Changes in AI Capabilities
9.1 Development and Improvement
Gneis Agency may:
- Upgrade to newer or better AI models
- Change or replace third-party providers
- Adjust algorithms and scoring methodologies
- Add new AI-driven features
9.2 Notice
Material changes affecting output quality or data protection are notified in accordance with Terms and Conditions.
9.3 Third-Party Limitations
If AI providers:
- Change their policies or terms
- Limit or discontinue their services
- Implement new restrictions
Gneis Agency may adjust the Services accordingly without this constituting a breach.
10. Customer Rights and Control
10.1 Transparency and Insight
The Customer has the right to:
- Information about which AI models are used
- Explanation of how outputs are generated (at high level)
- Access to this AI Transparency Policy
- Ask questions about AI use
10.2 Control and Preferences
Where technically possible, the Customer can:
- Choose between different AI models (Enterprise)
- Adjust detail and confidence levels
- Opt-out of certain AI features
- Request human review of critical outputs
10.3 Data Rights
The Customer retains:
- Ownership of input data (Customer Content)
- Right to export data (cf. EU Data Act)
- Right to deletion upon termination
- Right to restrict processing where relevant
11. AI-Related Support
11.1 Questions About AI
For questions about AI use, models or outputs:
ai@gneis.ai
11.2 Error Reporting
To report:
- Incorrect or biased outputs
- Security issues or abuse
- Unexpected AI behavior
support@gneis.ai or security@gneis.ai
11.3 Ethical Concerns
To report ethical concerns about AI use:
ethics@gneis.ai
12. Future AI Developments
12.1 Innovation and Research
Gneis Agency continuously invests in:
- Development of proprietary AI models
- Fine-tuning existing models for our domain
- Research in AI-driven marketing analytics
- Partnerships with leading AI research environments
12.2 Emerging Technologies
We monitor and evaluate:
- Multimodal AI (text, image, video, audio)
- Retrieval-Augmented Generation (RAG)
- Agentic AI and autonomous systems
- Federated learning and privacy-preserving AI
12.3 Commitment to Responsible Innovation
We commit to:
- Adopting new AI technologies responsibly
- Prioritizing safety and ethics over speed
- Consulting with experts and stakeholders
- Updating this policy as AI evolves
13. Regulatory Compliance
13.1 Applicable Regulation
This AI policy is designed to comply with:
- EU AI Act (Regulation (EU) 2024/1689)
- GDPR (Regulation (EU) 2016/679)
- EU Data Act (Regulation (EU) 2023/2854)
- Digital Services Act (Regulation (EU) 2022/2065)
- Copyright Directive (Directive (EU) 2019/790)
13.2 Ongoing Adaptation
We monitor regulatory developments and adapt our practices accordingly, including:
- Upcoming AI Act implementing acts
- National AI laws and guidelines
- Industry-specific regulations
- International standards (ISO/IEC 42001, etc.)
14. Contact and Further Information
For more information about AI use in Gneis Agency's Services:
Gneis Agency
Pilestræde 52A
DK-1112 Copenhagen K
Denmark
General: info@gneis.ai
AI questions: ai@gneis.ai
Privacy: privacy@gneis.ai
Security: security@gneis.ai
Ethics: ethics@gneis.ai
Related Documents:
- Terms and Conditions: https://trust.gneis.io/terms
- Privacy Policy: https://trust.gneis.io/privacy
- Acceptable Use Policy: https://trust.gneis.io/aup
- Data Processing Agreement: https://trust.gneis.io/dpa
*This AI Transparency & Use Policy is part of Gneis Agency's commitment to responsible and transparent AI use.*
