AI Terms of Use

These AI Terms of Use (“AI Terms”) supplement the Datafi Terms of Service (“Terms of Service”) and govern your access to and use of artificial intelligence and machine learning features available within the Datafi platform (the “AI Features”). These AI Terms are entered into by and between you (whether individually or on behalf of an entity, “you” or “Customer”) and Datafi Labs, Inc., a Delaware corporation (“Datafi,” “we,” “our,” or “us”).

By accessing or using any AI Features, you acknowledge that you have read, understood, and agree to be bound by these AI Terms in addition to the Terms of Service and our Privacy Policy. In the event of a conflict between these AI Terms and the Terms of Service, these AI Terms shall control with respect to the AI Features. Capitalized terms not defined herein shall have the meanings set forth in the Terms of Service.

1. Scope and Applicability

These AI Terms apply to all AI Features made available through the Datafi platform, whether integrated directly into existing Services such as Studio, Control Tower, Cyber, and Runtime, or offered as standalone capabilities. These AI Terms apply to all Authorized Users who interact with AI Features under Customer’s account.

If Customer has executed an Order Form or enterprise agreement that includes specific terms governing AI Features, those terms shall take precedence over these AI Terms to the extent they conflict.

2. Definitions

“AI Features” means any artificial intelligence, machine learning, natural language processing, or automated reasoning capabilities made available through the Datafi platform, including but not limited to data analysis suggestions, automated insights, natural language querying, anomaly detection, predictive analytics, and AI-assisted data transformations.
“AI Outputs” means any content, data, predictions, recommendations, summaries, visualizations, code, text, or other materials generated by the AI Features in response to Inputs.
“Inputs” means any data, prompts, queries, instructions, configurations, or other information that Customer or its Authorized Users submit to the AI Features for processing.
“Models” means the machine learning models, algorithms, neural networks, and related software components that power the AI Features, whether developed by Datafi or provided by Third-Party AI Providers.
“Training Data” means data used to train, fine-tune, validate, or improve Models.
“Third-Party AI Providers” means external providers of AI models, APIs, or related services that Datafi integrates with or utilizes in the delivery of AI Features.
“Feedback” means any suggestions, ratings, corrections, annotations, or other feedback that Customer or its Authorized Users provide regarding the AI Features or AI Outputs.

3. AI Feature Description

Datafi’s AI Features are designed to enhance the platform’s data management, analytics, and governance capabilities. AI Features may include, without limitation:

Natural language querying: enabling Authorized Users to interact with data using natural language prompts rather than structured query languages.
Automated insights and analytics: surfacing patterns, trends, and anomalies within Customer Data to support data-driven decision-making.
Predictive analytics: generating forecasts and predictive models based on historical data and statistical methods.
AI-assisted data transformations: suggesting or automating data cleaning, enrichment, mapping, and transformation workflows.
Anomaly detection: automatically identifying unusual patterns or outliers in data streams through Cyber.
Intelligent recommendations: providing recommendations for data governance policies, access controls, schema optimizations, and workflow configurations.
Summarization and reporting: generating natural language summaries, reports, and explanations from structured and unstructured data.
AI agents: autonomous or semi-autonomous software agents that can plan and execute multi-step tasks on behalf of an Authorized User, including querying and retrieving Customer Data, initiating data transformations or workflows, invoking Platform functions and connected integrations, and taking other actions within the scope of permissions granted by Customer, subject to the access controls, approval workflows, and governance policies configured by Customer within the Platform.

The specific AI Features available to Customer depend on the subscription plan and Order Form. Datafi may introduce, modify, or discontinue specific AI Features in accordance with Section 12 of these AI Terms.

4. AI Agents

4.1 Customer Responsibility for Agent Permissions

Where AI Features include AI agents capable of taking autonomous or semi-autonomous action, Customer is solely responsible for configuring the permissions, access controls, and approval workflows that govern what actions an AI agent may take on Customer’s behalf, including which Customer Data an agent may access, which Platform functions or connected integrations it may invoke, and whether human approval is required before an agent executes a given action. Datafi provides the controls; Customer determines how they are applied. Customer shall ensure that agent permissions are configured consistent with the principle of least privilege and are periodically reviewed, particularly following changes in personnel, Authorized User roles, or connected integrations. Datafi is not responsible for actions taken by an AI agent that are within the scope of permissions Customer has configured, even where Customer did not anticipate the specific action taken.

4.2 Agent Action Logging and Audit Trail

Datafi will log AI agent actions taken within the Platform, including the action taken, the Authorized User or workflow that initiated it, the Customer Data or systems affected, and the timestamp of execution, through Cyber. These logs form part of the audit trail available to Customer for governance, compliance, and incident-review purposes, consistent with Section 11 (Transparency and Explainability). Customer is responsible for monitoring agent activity logs relevant to its use case and for promptly notifying Datafi of any agent action it believes was taken outside the scope of configured permissions.

5. Customer Data and AI Training

Datafi does not use Customer Data to train, fine-tune, or improve its general-purpose Models or the Models of any Third-Party AI Provider without Customer’s explicit, prior, written opt-in consent. This commitment is fundamental to our approach to AI and data privacy.

5.2 Opt-In Training Programs

Datafi may offer optional programs that allow Customers to contribute anonymized or aggregated data toward improving AI Features. Participation in any such program is entirely voluntary, requires affirmative opt-in through a separate consent mechanism, and may be revoked by Customer at any time. Details regarding any opt-in training programs, including the specific data involved and its handling, will be communicated separately.

5.3 Processing for Service Delivery

To provide the AI Features, Datafi must process Inputs and Customer Data in real time. This processing is performed solely for the purpose of generating AI Outputs in response to Customer’s requests and does not constitute training. Transient data processed during inference is not retained beyond what is necessary to deliver the requested AI Output, subject to Section 10 of these AI Terms.

5.4 Feedback

If Customer provides Feedback regarding AI Features or AI Outputs, Datafi may use such Feedback to improve the AI Features and the Services generally. Feedback should not contain Customer Data or personally identifiable information. By providing Feedback, Customer grants Datafi a non-exclusive, worldwide, royalty-free, perpetual license to use, modify, and incorporate such Feedback into the Services.

6. Input and Output Ownership

6.1 Customer Inputs

Customer retains all right, title, and interest in and to its Inputs, including any Customer Data submitted to the AI Features. Nothing in these AI Terms transfers ownership of Customer’s Inputs to Datafi.

6.2 AI Outputs

As between Customer and Datafi, AI Outputs generated for Customer in response to Customer’s Inputs belong to Customer, subject to the following conditions:

Customer is responsible for reviewing, validating, and determining the suitability of AI Outputs before relying on them or incorporating them into any downstream processes, products, or decisions.
AI Outputs may not be unique. Other customers submitting similar Inputs may receive similar or identical AI Outputs. Datafi does not guarantee the exclusivity of any AI Output.
Customer’s rights in AI Outputs are subject to any applicable intellectual property rights of Datafi, its licensors, or Third-Party AI Providers in the underlying Models, algorithms, and technology used to generate such Outputs.
Customer’s use of AI Outputs must comply with all applicable laws, regulations, and the Terms of Service.

6.3 Datafi Intellectual Property

Nothing in these AI Terms grants Customer any right, title, or interest in the Models, algorithms, training methodologies, or other technology underlying the AI Features. Datafi and its licensors retain all intellectual property rights in the AI Features, Models, and the platform infrastructure.

7. Accuracy and Limitations

7.1 No Guarantee of Accuracy

AI Outputs are generated using probabilistic models and are not guaranteed to be accurate, complete, current, or error-free. AI technology, by its nature, may produce results that appear plausible but are factually incorrect, incomplete, or misleading. Datafi does not warrant that AI Outputs will meet Customer’s specific requirements or expectations.

7.2 Decision Support Only

AI Features and AI Outputs are provided as decision-support tools and are intended to assist, not replace, human judgment. Customer acknowledges and agrees that:

AI Outputs should be treated as suggestions, recommendations, or starting points that require human review and validation.
Customer must not rely solely on AI Outputs for any decision that could materially affect individuals, business operations, legal compliance, financial outcomes, or safety-critical systems.
Customer is solely responsible for verifying the accuracy, legality, and appropriateness of any AI Output before acting on it.

7.3 Known Limitations

Customer acknowledges that AI Features may have inherent limitations, including but not limited to potential biases in training data, sensitivity to the phrasing or structure of Inputs, difficulty with ambiguous or context-dependent queries, and reduced performance on data types or domains that are underrepresented in Model training. For more information on how Datafi addresses these limitations, please refer to our Responsible AI Framework.

8. Prohibited AI Uses

In addition to the prohibited uses set forth in the Terms of Service, Customer and its Authorized Users shall not use the AI Features to:

Generate harmful content: create, produce, or disseminate content that is defamatory, discriminatory, harassing, threatening, sexually explicit, or that promotes violence, illegal activities, or hatred against any individual or group.
Make autonomous high-stakes decisions without human oversight: use AI Outputs as the sole basis for decisions that have significant legal, financial, medical, safety, or civil rights implications for individuals, including but not limited to employment decisions, credit determinations, criminal justice assessments, healthcare diagnoses, or insurance underwriting.
Circumvent safety measures: attempt to bypass, disable, override, or manipulate any safety filters, content moderation systems, guardrails, rate limits, or usage restrictions implemented in the AI Features.
Reverse engineer Models: attempt to extract, reconstruct, distill, or reverse engineer the underlying Models, weights, training data, or algorithms through systematic querying, model probing, or any other technique.
Generate deceptive content: create synthetic data, deepfakes, or other AI-generated content intended to deceive, defraud, or mislead individuals or organizations.
Conduct unauthorized surveillance: use AI Features to monitor, profile, or track individuals without appropriate legal authority and consent.
Violate intellectual property rights: use AI Features to intentionally generate content that infringes on the intellectual property rights of third parties.
Develop competing products: use AI Features, AI Outputs, or any information derived therefrom to develop, train, or improve AI models or services that compete with Datafi.

Datafi reserves the right to suspend or terminate access to AI Features for any Customer or Authorized User who violates these prohibited use provisions, without liability and without prejudice to any other remedies available under the Terms of Service.

9. Third-Party AI Models and Providers

9.1 Use of Third-Party Providers

Datafi may integrate with or utilize Third-Party AI Providers to deliver certain AI Features. Datafi selects Third-Party AI Providers through a rigorous evaluation process that includes assessments of security practices, data handling policies, performance, and reliability.

When Inputs or Customer Data are processed by Third-Party AI Providers, Datafi ensures that such processing is governed by contractual agreements that require the Third-Party AI Provider to: (a) process data solely for the purpose of delivering the requested AI Output; (b) refrain from using Customer Data to train its own models, except where Customer has provided separate, explicit consent; (c) implement appropriate technical and organizational security measures; and (d) comply with applicable data protection laws.

9.3 Third-Party Terms

Certain Third-Party AI Providers may impose additional terms or usage policies. Datafi will make reasonable efforts to notify Customers of material third-party terms that affect their use of AI Features. Customer agrees to comply with any such third-party terms as communicated by Datafi.

9.4 Availability

The availability of AI Features that depend on Third-Party AI Providers is subject to the continued availability of those third-party services. Datafi shall not be liable for any interruption, degradation, or discontinuation of AI Features caused by a Third-Party AI Provider.

10. Data Handling for AI Features

10.1 Processing

Inputs submitted to AI Features are processed in order to generate AI Outputs. Processing may occur in Datafi’s own infrastructure or through the infrastructure of Third-Party AI Providers, subject to the protections described in Section 9. Datafi processes Inputs in accordance with our Privacy Policy and any applicable data processing agreement.

10.2 Retention

Datafi retains Inputs and AI Outputs only for as long as necessary to: (a) provide and maintain the AI Features; (b) fulfill Customer’s requests, including maintaining conversation history or audit logs where applicable; (c) comply with legal obligations; and (d) resolve disputes or enforce agreements. Specific retention periods may vary by AI Feature and are documented in the applicable product documentation.

10.3 Deletion

Customer may request deletion of Inputs and AI Outputs stored by Datafi in accordance with the data deletion provisions of the Terms of Service and Privacy Policy. Upon termination of Customer’s subscription, Datafi will delete or return Customer Data, including Inputs and AI Outputs, in accordance with the Terms of Service, subject to any legal or regulatory retention requirements.

10.4 Logging and Monitoring

Datafi may log metadata related to AI Feature usage (such as timestamps, feature identifiers, token counts, and error codes) for purposes of service operation, performance monitoring, security, abuse prevention, and billing. Such logs do not contain the substantive content of Inputs or AI Outputs unless required for debugging with Customer’s consent.

11. Transparency and Explainability

11.1 Identification of AI-Generated Content

Datafi is committed to transparency regarding the use of AI within the platform. AI Features are clearly identified within the user interface so that Authorized Users can distinguish between AI-generated content and other platform outputs. For more information on our transparency commitments, see our Responsible AI Framework.

11.2 Explainability

Where technically feasible, Datafi strives to provide explanations for AI Outputs, including the reasoning or data points that contributed to a particular result. The level of explainability may vary depending on the complexity of the underlying Model, the nature of the AI Feature, and technical constraints.

11.3 Documentation

Datafi maintains documentation that describes the general capabilities, intended use cases, and known limitations of each AI Feature. This documentation is available through the platform’s help center and is updated as AI Features evolve.

11.4 Bias Mitigation

Datafi employs measures to identify, assess, and mitigate potential biases in AI Features and AI Outputs. While no AI system can be entirely free from bias, Datafi is committed to continuous monitoring and improvement. Details about our bias mitigation practices are available in our Responsible AI Framework.

12. Updates to AI Features

12.1 Modifications

Datafi may update, modify, enhance, or discontinue AI Features at any time. Such changes may include updating underlying Models, adjusting safety filters, adding new capabilities, or removing features that no longer meet our quality or safety standards.

12.2 Notice

For material changes to AI Features that are reasonably likely to affect Customer’s use, Datafi will provide advance notice through the platform, email, or other reasonable communication channels. The notice period will be proportionate to the significance of the change, with a minimum of thirty (30) days for changes that materially reduce the functionality of an AI Feature included in Customer’s subscription.

12.3 Model Updates

Models powering AI Features may be updated or replaced to improve performance, safety, or accuracy. Such updates may result in changes to AI Outputs, even for identical Inputs. Datafi does not guarantee consistency of AI Outputs across Model updates.

13. Liability for AI Outputs

13.1 No Warranty for AI Outputs

AI Features and AI Outputs are provided “as is” and “as available.” To the maximum extent permitted by applicable law, Datafi disclaims all warranties, express or implied, with respect to AI Outputs, including warranties of accuracy, fitness for a particular purpose, merchantability, and non-infringement.

13.2 Limitation of Liability

Datafi’s liability for claims arising from or related to the AI Features or AI Outputs is subject to the limitation of liability provisions in the Terms of Service. In no event shall Datafi be liable for any damages arising from Customer’s reliance on AI Outputs without appropriate human review and validation, including but not limited to decisions made based on inaccurate, incomplete, or biased AI Outputs.

13.3 Customer Responsibility

Customer is solely responsible for: (a) determining whether AI Features and AI Outputs are appropriate for Customer’s intended use cases; (b) implementing adequate human oversight and review processes for AI Outputs; (c) ensuring that its use of AI Features and AI Outputs complies with all applicable laws, regulations, and industry standards; and (d) any consequences arising from Customer’s use of or reliance on AI Outputs.

13.4 Indemnification

Customer shall indemnify, defend, and hold harmless Datafi and its officers, directors, employees, and agents from and against any claims, damages, losses, liabilities, costs, and expenses (including reasonable attorneys’ fees) arising from or related to: (a) Customer’s use of AI Features in violation of these AI Terms, the Terms of Service, or applicable law; (b) Customer’s reliance on AI Outputs without appropriate human oversight; or (c) any content generated through Customer’s use of AI Features that infringes third-party rights.

14. Contact

If you have questions, concerns, or requests regarding these AI Terms, Datafi’s AI Features, or our AI practices, please contact us at:

Datafi Labs, Inc. 314 Williams Ave South #1245, Renton, WA 98057, United States Email: [email protected]

For matters related to data privacy and AI data handling, you may also refer to our Privacy Policy. For information about our approach to responsible AI development and deployment, please see our Responsible AI Framework.