Provision Registry

203 classified provisions across 105 platforms — browse, filter, and compare.

Every clause classified by type, severity, and platform. Updated as policies change.

Track provision changes Get alerts when clauses you care about change across platforms
Filtering: Liability limitation × Clear all
medium Liability limitation
Google · Google AI Principles
Accountability and explainability are increasingly legal requirements — particularly under GDPR Article 22 for automated decision-making — and this commitment establishes Google's public standard for …
CA-P-002368 First tracked Apr 9, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
Character.AI · Character.AI Terms of Service
The company takes no responsibility for the accuracy, safety, or appropriateness of AI outputs, which is significant given the platform's potential use by vulnerable users …
CA-P-000802 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
Microsoft · Responsible AI
AI reliability failures in safety-critical applications like healthcare diagnosis, autonomous vehicles, or infrastructure control can cause physical harm, making this commitment directly relevant to consumer …
CA-P-002519 First tracked Apr 9, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
Netflix · Netflix Account and Content Policies
This disclaimer means that if Netflix goes down, content becomes unavailable, or the service malfunctions, you have no contractual right to compensation — though statutory …
CA-P-002194 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Liability limitation
Midjourney · Midjourney Terms of Service
Users have no contractual basis to seek compensation from Midjourney if the AI generates poor-quality, inaccurate, or inappropriate content, and Midjourney caps its total liability …
CA-P-000685 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Netflix · Netflix Account and Content Policies
If Netflix's service fails or causes you a loss, your ability to seek compensation is severely limited, even if Netflix's actions caused harm beyond just …
CA-P-000359 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
Headspace · Headspace Terms and Conditions
On a platform that also offers therapy and psychiatry services through affiliated providers, the line between what constitutes professional medical advice and general wellness content …
CA-P-001128 First tracked Apr 3, 2026 Last seen Apr 27, 2026 Compare across platforms →
Anthropic · Anthropic Privacy Policy
Millions of enterprise users interacting with Claude through their employers have no direct privacy relationship with Anthropic and must rely entirely on their employer's privacy …
CA-P-002563 First tracked Apr 9, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Liability limitation
Salesforce · Salesforce Terms of Service
These agreements are the legally binding contracts between Salesforce and its business customers — they define what Salesforce must deliver, what customers can do with …
CA-P-001083 First tracked Apr 3, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
Noom · Noom Privacy Policy
The 'reasonable measures' standard is vague and provides no specific security guarantee — if a breach occurs, this language limits Noom's accountability to users.
CA-P-001850 First tracked Apr 3, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
LinkedIn · LinkedIn User Agreement
This means that if LinkedIn experiences outages, provides inaccurate job listings, or surfaces misleading professional information, you generally have no contractual right to compensation.
CA-P-002161 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Liability limitation
Figma · Figma Terms of Service
Figma expressly disclaims all warranties, so if the platform experiences an outage, data loss, or security breach, you cannot hold Figma responsible for failing to …
CA-P-001106 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
GitHub · GitHub Terms of Service
GitHub makes no promises about service availability or reliability, meaning if the platform goes down and you lose work or suffer business disruption, GitHub has …
CA-P-001341 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
Strava · Strava Terms of Service
If the Strava app fails during a workout, produces inaccurate health data, or experiences a security incident, you cannot hold Strava responsible based on any …
CA-P-001427 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Skillshare · Skillshare Terms of Service
If Skillshare's service fails, loses your data, or causes you harm, your ability to recover compensation is severely restricted regardless of the severity of the …
CA-P-001289 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
Google · Google Terms of Service
Google's 'as is' disclaimer means you have no contractual right to a working, reliable service — if Gmail goes down and you lose important emails …
CA-P-001898 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Liability limitation
Google · Google AI Principles
This is one of the most specific and unusual restrictions published by any major technology company, creating a public commitment that constrains government contract opportunities …
CA-P-002369 First tracked Apr 9, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
Google · Google AI Principles
Algorithmic bias in AI systems used for consequential decisions (search rankings, loan approvals, healthcare recommendations) can cause real-world harm to protected groups — this commitment …
CA-P-002365 First tracked Apr 9, 2026 Last seen Apr 27, 2026 Compare across platforms →
medium Liability limitation
Microsoft · Microsoft Responsible AI Principles
Discriminatory AI outcomes in lending, hiring, or healthcare can cause serious real-world harm; this principle creates an expectation but no enforcement mechanism for affected individuals.
CA-P-002529 First tracked Apr 9, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
Peloton · Peloton Terms of Service
If you are injured while using Peloton equipment or following its fitness programs, Peloton disclaims responsibility, placing full accountability on you as the user.
CA-P-001171 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
Spotify · Spotify Terms and Conditions
This limits your ability to recover compensation for service outages even if you are a paying subscriber who loses access to a paid service, as …
CA-P-002172 First tracked Apr 4, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Liability limitation
Anthropic · Anthropic Usage Policy
This carve-out allows government customers to operate outside the standard AUP restrictions, creating a two-tiered system where activities prohibited for commercial users may be permitted …
CA-P-002572 First tracked Apr 9, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
Noom · Noom Terms of Service
Users dealing with health conditions who rely on Noom's program as health guidance have no legal recourse against Noom if that guidance is incorrect or …
CA-P-001838 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
Amazon · Amazon Conditions of Use
Unauthorized use of Amazon's copyrighted content — including screen-scraping, automated data collection, or reproducing product listings — can expose users and developers to copyright infringement …
CA-P-002538 First tracked Apr 9, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
OpenAI · Terms of Use (ROW)
AI-generated outputs may lack copyright protectability under current US Copyright Office guidance, and the non-uniqueness warning means you could publish or commercially use content that …
CA-P-002446 First tracked Apr 9, 2026 Last seen Apr 10, 2026 Compare across platforms →
medium Liability limitation
Webull · Webull Customer Agreement
Webull makes no representations about the suitability of any investment for your personal financial situation, meaning users should not rely on the platform for personalized …
CA-P-000489 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
PayPal · PayPal User Agreement
Closing your account does not end your financial exposure to PayPal — you can still be pursued for money owed, disputed transactions, or chargeback losses …
CA-P-002283 First tracked Apr 9, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Liability limitation
BeReal · BeReal Terms of Service
If BeReal's platform causes you harm — such as a data breach or wrongful account deletion — your ability to seek compensation is severely limited …
CA-P-001304 First tracked Apr 3, 2026 Last seen Apr 17, 2026 Compare across platforms →
medium Liability limitation
PayPal · PayPal User Agreement
Closing your account does not end your financial exposure to PayPal — outstanding chargebacks, disputed transactions, or fees can still be collected from you after …
CA-P-002257 First tracked Apr 5, 2026 Last seen Apr 9, 2026 Compare across platforms →
medium Liability limitation
Messenger Kids · Messenger Kids Terms of Service
If your child is harmed through the service — for example, through exposure to inappropriate content or a data breach — Meta's financial liability to …
CA-P-000584 First tracked Apr 3, 2026 Last seen Apr 10, 2026 Compare across platforms →

Don't read every policy manually.

Get alerts when the clauses that matter to you change across 105 platforms.

Watcher — $9.99/mo Professional — $149/mo