Cross-platform comparison

AI / Automated Decision-Making, Cross-Platform Comparison

217 platforms tracked · 922 provisions · 174 platforms high severity · Compare how platforms differ on this clause type.

217
Platforms
174
High severity
145
Medium
28
Low
352
Total monitored
Stay ahead of the changes

Track clause changes

Get alerts when any platform updates this clause.

Compare platforms side by side
Select 2 to 4 platforms. Comparison shows clause text, consumer impact, and AI-generated difference analysis.
Platforms with AI / Automated Decision-Making provisions Click to select for comparison
Acorns financial
No automated decision-making affecting users personally
✓ Clause text available High
ActiveCampaign platform
No Warranty That AI Output Does Not Infringe Third Party Rights
✓ Clause text available High
Adobe infrastructure
Prohibition on Using Services Output to Train AI or ML Systems
✓ Clause text available High
ADP financial
EEA automated decision-making notification right
✓ Clause text available High
Affirm financial
AI Tool Transaction Treated as Authorized by User
✓ Clause text available High
Afterpay financial
Prohibition on Using AI Outputs to Train Models
✓ Clause text available High
AI21 Labs ai
AI Processing Personal Data via AI Systems
✓ Clause text available High
Amazon gatekeeper
Agent Prohibited from Mimicking Human Behavior
✓ Clause text available High
Amazon Associates gatekeeper
Agents Must Identify Themselves in HTTP Requests
✓ Clause text available High
American Airlines infrastructure
Brazil Right to Oppose Automated Decision-Making
✓ Clause text available High
Ancestry education
Face grouping using abstract image representations with consent
✓ Clause text available High
Anthropic ai
Automated R&D Capability Threshold Revision
✓ Clause text available High
Apple App Store platform
Private personal data not used to train AI models
✓ Clause text available High
Asana platform
AI user responsible for all decisions from AI use
✓ Clause text available High
Atlassian developer_tools
AI offerings governed by AI terms
✓ Clause text available High
AWS infrastructure
AWS does not engage in legally significant automated profiling
✓ Clause text available High
AWS Bedrock infrastructure
Connect Talent bias monitoring obligation
✓ Clause text available High
BeReal social
Interest inference from visual elements of photos
✓ Clause text available High
Best Buy marketplace
AI statements on Terms are void and non-binding
✓ Clause text available High
Betterment financial
AI Provider Not to Use Data for Model Training
✓ Clause text available High
Binance.US financial
BAM Does Not Use Automated Decision-Making Impacting Employees
✓ Clause text available High
Booking.com marketplace
AI-powered assistants prohibited from making bookings without permission
✓ Clause text available High
Box platform
User solely responsible for Box AI output retention and compliance
✓ Clause text available High
Brex financial
AI and Automated Fraud Detection Processing
✓ Clause text available High
Bumble platform
Automated Profiling and Matching Algorithms
✓ Clause text available High
Calendly platform
Customer Input and Output shared with third-party AI providers
✓ Clause text available High
Canva infrastructure
No AI Training on Canva Education Content
✓ Clause text available High
Cash App financial
AI Features Do Not Create Fiduciary Duty
✓ Clause text available High
Chegg education
User Solely Responsible for AI Output Third-Party Rights
✓ Clause text available High
Cohere ai
Prohibition on Automated Decisions Affecting Individual Rights
✓ Clause text available High
Coinbase financial
Automated decision-making for safety and security obligations
✓ Clause text available High
Comcast gatekeeper
AI Agent Definition Includes Autonomous Software
✓ Clause text available High
Copy.ai ai
Right Against Solely Automated Decision-Making
✓ Clause text available High
Coursera education
Right to elect human grader for AI peer review
✓ Clause text available High
Cursor ai
No Automated Decision-Making on Users
✓ Clause text available High
Datadog developer_tools
OpenAI AI services subprocessor
✓ Clause text available High
DeepL platform
Customer Responsible for High-Risk EU AI Act Compliance
✓ Clause text available High
DeepSeek ai
Fully Automated Adverse Legal Decision Prohibition
✓ Clause text available High
Descript ai
AI Output May Be Inaccurate or Infringing
✓ Clause text available High
Discord social
Content used to train automated moderation systems
✓ Clause text available High
Disney+ media
Prohibition on AI Tool Use of Content
✓ Clause text available High
DocuSign infrastructure
AI Models Designed to Avoid Personal Information Training
✓ Clause text available High
DoorDash marketplace
Prohibition on AI Training Using Service Content
✓ Clause text available High
Dun & Bradstreet infrastructure
AI systems use in data processing
✓ Clause text available High
eBay marketplace
Automated Decision-Making Restriction Without Basis
✓ Clause text available High
Epic Games gaming
Machine learning analysis of user communications
✓ Clause text available High
Eufy infrastructure
User Solely Responsible for Verifying AI-Generated Output
✓ Clause text available High
Faire marketplace
Automated Invoice Terms Determination Using Retailer Data
✓ Clause text available High
Figma platform
Children Personal Information AI Training Prohibition
✓ Clause text available High
General Motors infrastructure
AI assistant interaction information collection
✓ Clause text available High
GitHub developer_tools
User Must Review and Validate AI Output Before Use
✓ Clause text available High
Glassdoor platform
No Intellectual Property Rights in AI Offerings or Models
✓ Clause text available High
Glean ai
LGPD Right to Review Automated Decision-Making
✓ Clause text available High
GOAT marketplace
Prohibition on AI-Generated Fake or Misleading Content
✓ Clause text available High
Google gatekeeper
Prohibition on Training AI with Google AI Content
✓ Clause text available High
Google Cloud platform
Google may log Generative AI prompts to detect abuse
✓ Clause text available High
Google DeepMind ai
Acknowledgment Automated Monitoring Long-Term Insufficiency
✓ Clause text available High
Google Gemini ai
Restriction on Unsupervised High-Risk Automated Decisions
✓ Clause text available High
Google Maps gatekeeper
Prohibition on Using Maps Content for AI Model Training
✓ Clause text available High
Grammarly platform
AI Output Factual Accuracy Disclaimer
✓ Clause text available High
Greenhouse platform
Mandatory Third-Party Bias Audit Requirement
✓ Clause text available High
Grindr platform
User content license includes AI development use
✓ Clause text available High
Groq ai
No Solely Automated Decision-Making Without Legal Ground
✓ Clause text available High
Gusto financial
User solely responsible for reviewing AI outputs
✓ Clause text available High
Harvey AI ai
Harvey Will Not Train AI Models on Customer Content or Data
✓ Clause text available High
Headspace health
AI Feature User Notification Requirement
✓ Clause text available High
Hilton financial
Automated Decisions Made With Consent for Special Offers
✓ Clause text available High
Hims & Hers health
AI Not Used For Clinical Decisions Or Mental Health Care
✓ Clause text available High
Hinge platform
Profiling and Automated Decision-Making Use
✓ Clause text available High
Home Depot marketplace
Enhanced Facial Verification for Fraud Prevention
✓ Clause text available High
HubSpot platform
Google Workspace APIs not used for generalized AI/ML training
✓ Clause text available High
Hulu media
Prohibition on Using Content for AI Tool Development
✓ Clause text available High
Indeed platform
Apply For Me Tool Auto-Submits Applications Based on Profile
✓ Clause text available High
Instacart marketplace
Prohibition on Harmful or Deceptive Automated System Use
✓ Clause text available High
Intuit financial
Personal data used to train AI and ML models
✓ Clause text available High
Kajabi platform
Prohibition On Using MCP Data To Train AI
✓ Clause text available High
Khan Academy education
User Solely Responsible for AI Input and Output Use
✓ Clause text available High
Klaviyo platform
Customer solely responsible for AI Inputs and Outputs
✓ Clause text available High
Leonardo AI ai
Content Used for Machine Learning and Model Training
✓ Clause text available High
Linear platform
AI Services fees governed by AI Addendum
✓ Clause text available High
LinkedIn social
User Responsible for AI-Generated Content Shared
✓ Clause text available High
Loom platform
AI Offerings governed by AI Terms
✓ Clause text available High
Luma AI ai
Customer Authorization for Third-Party AI Tool Actions
✓ Clause text available High
Lyft marketplace
Automated processes use personal information for platform operation
✓ Clause text available High
Mailchimp platform
Member Responsible for Reviewing AI Outputs Before Use
✓ Clause text available High
McDonald's marketplace
McDonald's uses data to train AI and algorithms
✓ Clause text available High
Mercury financial
AI human oversight for consequential decisions
✓ Clause text available High
MetaMask financial
No Consensys responsibility for AI Agent compromise
✓ Clause text available High
Microsoft gatekeeper
Prohibition on consequential AI decisions without human oversight
✓ Clause text available High
Microsoft Copilot gatekeeper
AI prohibition on inferring emotional states from physical characteristics
✓ Clause text available High
Mistral AI ai
Customer Cooperation on Systemic Risk Mitigation
✓ Clause text available High
Mixpanel platform
Personal data used for profiling future interest
✓ Clause text available High
Modal infrastructure
Modal Prohibited from Training AI on Customer Data
✓ Clause text available High
Neon platform
Use of AI and large language models to process data
✓ Clause text available High
Netflix media
Prohibition on AI Training Use of Service
✓ Clause text available High
Nextdoor social
Automated and manual systems analyze user content
✓ Clause text available High
Noom health
Prohibition On Using Service To Train AI Models
✓ Clause text available High
Notion developer_tools
Workspace API data not used for AI training
✓ Clause text available High
Okta platform
Right not to be subject to automated decisions
✓ Clause text available High
OnlyFans media
No Solely Automated Decision-Making with Legal Effect
✓ Clause text available High
OpenAI ai
Model Training Data Use
✓ Clause text available High
Oscar Health health
AI output susceptible to inaccuracies and hallucinations
✓ Clause text available High
Oura health
Oura may use personal data to train its own AI and LLM features
✓ Clause text available High
Patreon platform
Patreon makes inferences including demographic and earnings estimates
✓ Clause text available High
PayPal financial
AI Model Training Using Personal Information
✓ Clause text available High
Peacock gaming
Prohibition on Using Content to Train AI Models
✓ Clause text available High
Peloton health
Third-Party AI Providers Receive Only Deidentified Data
✓ Clause text available High
Perplexity AI ai
Customer Responsible for Verifying Output Accuracy
✓ Clause text available High
Pinterest social
User data used to train machine learning models
✓ Clause text available High
PlanetScale platform
Customer responsible for all AI input and output
✓ Clause text available High
Postman developer_tools
Use of Data to Develop AI Tools
✓ Clause text available High
Public.com financial
Non-US Alpha Users May Object to Automated Decisions
✓ Clause text available High
Ramp financial
No AI training on Services without written consent
✓ Clause text available High
RapidAPI platform
Prohibition on submitting personal data to chatbot
✓ Clause text available High
Reddit social
No automated profiling with legal or significant effects
✓ Clause text available High
Replicate ai
Models Not Intended to Provide Professional Advice
✓ Clause text available High
Replit ai
Public app content used for AI training
✓ Clause text available High
Revolut financial
Automated Decision-Making and Credit Eligibility
✓ Clause text available High
Ring infrastructure
No Guarantee AI Technology Accuracy Availability
✓ Clause text available High
Riot Games gaming
Automated Grading System for Inappropriate Behavior Consequences
✓ Clause text available High
Ro health
AI Processing of User Information for Business Purposes
✓ Clause text available High
Robinhood financial
Use data for AI training and enhancement
✓ Clause text available High
Roblox gaming
Brazil automated decision-making not used
✓ Clause text available High
Rumble media
Grant of Right to Use Content for AI Training
✓ Clause text available High
RunPod infrastructure
No Automated Decision-Making for European Users
✓ Clause text available High
Salesforce platform
Right to object to automated decision-making granted
✓ Clause text available High
Samsung infrastructure
AI Output May Be Inaccurate Inappropriate or Infringing
✓ Clause text available High
Segment developer_tools
AI/ML model training for security and fraud prevention
✓ Clause text available High
SimpliSafe infrastructure
No machine learning profiling without consent
✓ Clause text available High
Skillshare media
EU and UK Right to Object to Automated Decision-Making
✓ Clause text available High
Smartsheet platform
AI Output As-Is No Professional Advice Disclaimer
✓ Clause text available High
Snapchat social
Snap Group AI Inputs Outputs treated as user content
✓ Clause text available High
SoFi financial
AI Coach responses may be incorrect and not relied upon for advice
✓ Clause text available High
Sourcegraph Cody ai
AI Terms apply to pre-February 2024 agreements
✓ Clause text available High
Spotify media
Spotify Does Not Use Automated Decision-Making
✓ Clause text available High
Square financial
Profile data used for automated identity verification
✓ Clause text available High
Stability AI ai
Ban on Social Scoring Systems
✓ Clause text available High
Starbucks marketplace
EEA Right Against Automated Decision-Making
✓ Clause text available High
Stash financial
Automated Profiling Models Built from Personal Data
✓ Clause text available High
StockX marketplace
EEA Automated Decision-Making and Profiling Disclosure
✓ Clause text available High
Strava health
Information used for AI and ML model training
✓ Clause text available High
Stripe financial
AI model training using Personal Data
✓ Clause text available High
Substack media
No final content moderation decision without human review
✓ Clause text available High
Supabase developer_tools
AI Input Must Not Violate Law or Third-Party Rights
✓ Clause text available High
Synthesia ai
Customer Responsible for Lawful AI Content Use
✓ Clause text available High
Target marketplace
AI Model Development and Training Using Data
✓ Clause text available High
Telegram social
Automated Algorithms Analyze Cloud Chats for Spam
✓ Clause text available High
Thomson Reuters infrastructure
Right to opt out of automated decision making and profiling
✓ Clause text available High
TikTok social
Prohibition On Circumventing AI Safety Controls
✓ Clause text available High
TikTok Ads social
Nigeria right to object to automated decision-making
✓ Clause text available High
T-Mobile gatekeeper
Personal data used to train artificial intelligence models
✓ Clause text available High
Twilio developer_tools
Automated security decisions with right to object
✓ Clause text available High
Uber marketplace
Facial Recognition to Verify Account User Identity
✓ Clause text available High
Unity gaming
AI Agent Acts or Omissions Attributed to User
✓ Clause text available High
Unreal Engine gaming
Prohibition on Using MetaHuman for AI Database Building
✓ Clause text available High
Vercel developer_tools
User Legally Bound by AI Autonomous Actions
✓ Clause text available High
Vercel AI ai
Third-Party AI Model Training Based on Legitimate Interests with Opt-Out
✓ Clause text available High
Verizon gatekeeper
Monitor and record customer communications with AI
✓ Clause text available High
Walmart marketplace
AI Output Not for Legal or Safety Decisions
✓ Clause text available High
Wealthfront gaming
Home lending automated calculations subject to human review
✓ Clause text available High
Weights & Biases ai
AI ML notice or consent required by law
✓ Clause text available High
Whoop health
User Content License for AI Model Training
✓ Clause text available High
Windsurf ai
User Content Used to Train AI Models
✓ Clause text available High
Workday platform
Recorded call content analyzed using AI-powered tools
✓ Clause text available High
Wyze infrastructure
Friendly Faces service infers individual identity from Facial Data
✓ Clause text available High
X social
Third-party collaborator data sharing and AI training
✓ Clause text available High
xAI ai
Prohibition On High-Stakes Automated Decisions
✓ Clause text available High
Xbox gaming
AI Prohibition on Biometric-Based Categorization of Beliefs
✓ Clause text available High
Yelp platform
AI Training Use of User Content
✓ Clause text available High
Zelle financial
Use of Behavioral Advertising Cookies On and Off Website
✓ Clause text available High
Zendesk platform
Zendesk to Notify if Automated Decision-Making Introduced
✓ Clause text available High
Zillow marketplace
Interest-Based Advertising via Third Parties
✓ Clause text available High
ZipRecruiter platform
Client ATS Screening Functionality Not Hiring Decision
✓ Clause text available High
Zoom infrastructure
Zoom Will Not Use Content to Train AI Models
✓ Clause text available High
Airbnb marketplace
Airbnb content recommendation systems use personalization criteria
✓ Clause text available Low
Fastly platform
Fastly does not use automated decision making
✓ Clause text available Low
Hugging Face ai
Trending content influenced by recent likes
✓ Clause text available Low
Inflection AI ai
IP Address Location Inference
✓ Clause text available Low
Paramount+ media
Paramount makes content recommendations based on usage
✓ Clause text available Low
Redfin platform
Redfin Does Not Use Profiling For Significant Decisions
✓ Clause text available Low
Activision gaming
Activision does not engage in legal-effect profiling
✓ Clause text available Medium
Airtable platform
AI and Automated Decision Making Features
✓ Clause text available Medium
Amplitude platform
Amplitude does not use data for automated decision-making
✓ Clause text available Medium
Bluesky social
Use of Hash-Matching and Text Detection Tools
✓ Clause text available Medium
Checkout.com financial
ML Fraud Score Reasons May Not Be Provided
✓ Clause text available Medium
Databricks infrastructure
AI Model Use Subject to Additional Terms
✓ Clause text available Medium
Dropbox platform
Automated technologies deployed to detect abusive content
✓ Clause text available Medium
Eventbrite platform
Expansion of inference capabilities over time
✓ Clause text available Medium
Ford infrastructure
Ford Infers User Preferences Using Automated Means
✓ Clause text available Medium
Gemini financial
Anonymized Data Used to Train AI Models
✓ Clause text available Medium
Google Ads gatekeeper
AI And Human Evaluation Enforces Compliance
✓ Clause text available Medium
Grubhub marketplace
SMS Messages May Use Automated Technology
✓ Clause text available Medium
Heap platform
Marketing profiling based on collected personal data
✓ Clause text available Medium
Kick platform
Profiling based on personal information for targeted content
✓ Clause text available Medium
LlamaIndex ai
LlamaIndex Does Not Train Models on User Content
✓ Clause text available Medium
Meta social
AI Products Subject To Meta AI Terms
✓ Clause text available Medium
Microsoft Azure platform
AI and manual processing of personal data
✓ Clause text available Medium
Miro platform
Automated Analysis of Service Use Cases
✓ Clause text available Medium
MyFitnessPal health
Use of AI and Machine Learning to Personalize Services
✓ Clause text available Medium
Reverb marketplace
Automated Fraud Detection Exempt from Opt-Out
✓ Clause text available Medium
Runway ai
AI outputs may not be unique across users
✓ Clause text available Medium
Salesforce Einstein ai
Dynamic Grounding Forces LLM to Use Approved Sources
✓ Clause text available Medium
Shein marketplace
SHEIN Claims No Automated Decision-Making Subject to Rights
✓ Clause text available Medium
Shopify infrastructure
Machine learning human involvement or low impact
✓ Clause text available Medium
Slack platform
Slack Uses Automated Processing To Prevent Abuse
✓ Clause text available Medium
Squarespace financial
Automated Processing Fraud Detection Exemption
✓ Clause text available Medium
Steam gaming
Valve May Use Automated Detection To Enforce Conduct Rules
✓ Clause text available Medium
Tabnine ai
EU/EEA right to object to automated decision-making
✓ Clause text available Medium
Teachable platform
Teachable Not Liable Under AI Provider Terms
✓ Clause text available Medium
Threads social
AI interaction data used to improve Meta AI
✓ Clause text available Medium
Ticketmaster marketplace
Global right to contest automated decisions
✓ Clause text available Medium
Tinder platform
User Responsible for AI-Generated Content They Share
✓ Clause text available Medium
Uniswap financial
Support Chatbot No Professional Advice
✓ Clause text available Medium
Walgreens marketplace
State law profiling via credit card prescreening
✓ Clause text available Medium
Writer ai
Independent toxicity classification system monitoring
✓ Clause text available Medium
Xfinity infrastructure
Automated Scanning of Communications for Security
✓ Clause text available Medium
YouTube Ads media
YouTube May Use Automated Systems to Analyze Content
✓ Clause text available Medium
Stay ahead of the changes

Full clause analysis

Get same-day alerts and full clause analysis for every tracked platform.

Browse by provision type