Live feed · updated daily
Recent policy changes
3515 policy changes
detected across 352 platforms. Most platforms don't announce policy changes — these updates were detected automatically.
July 31, 2026
Navigation restructure adds 'Products' and 'Solutions' menu items to Safety Policy page; no policy language changes.
Why it matters: This change has minimal operational significance. It reorganizes the website navigation on ElevenLabs' Safety Policy page by adding menu items but does not modify any policy terms, safety commitments, or user-facing obligations.
Navigation menu restructured; no policy or terms changes detected.
Why it matters: This change does not affect the substantive terms of service or user obligations. It is a navigation and formatting change only, with no operational impact on how the agreement applies or what parties are required to do.
Navigation menu restructured on ElevenLabs privacy policy page; substantive policy language unchanged.
Why it matters: This change does not alter the substantive terms of ElevenLabs' privacy policy. The update is limited to website navigation structure and has no operational significance for privacy rights or data processing practices.
Clarifies that English language version is legally binding; identifies DeepL SE as service provider.
Why it matters: The updated terms clarify that the English language version controls all contractual interpretation. This is operationally significant for users relying on non-English translations, as any discrepancies will be resolved in favor of the English text. The formal identification of DeepL SE establishes the legal entity responsible for performance and dispute resolution.
Navigation and footer restructuring; no substantive policy changes to user rights or obligations.
Why it matters: This change has no material operational significance. The reorganization of the document's navigation structure and footer does not modify substantive terms, user rights, data handling practices, or service obligations.
Stay ahead of the changes
You're seeing a fraction of what's changing
ConductAtlas monitors every tracked platform and captures every policy update.
Navigation menu reorganization; no substantive privacy policy changes detected.
Why it matters: This change does not materially affect privacy practices, data collection, or user rights. The reorganization is navigational only.
Privacy policy revision detected; specific language change not visible in available documentation.
Why it matters: Privacy policy modifications can materially affect what data is collected, how it is used, who it is shared with, and what rights users have over their information. Without visibility into the specific language that changed, the operational significance cannot be determined; a complete policy diff is needed.
Price reference updated from $1,908.39 to $1,920.00 in terms.
Why it matters: A price reference in the terms was adjusted, but without visible context describing what commercial term this figure represents or to whom it applies, the operational significance cannot be reliably assessed. Full document review is necessary to determine whether this change affects user fees, renewal costs, or other material commercial terms.
Removed cookie consent interface from Terms homepage; Cookies Policy remains available.
Why it matters: The cookie consent interface is a front-line mechanism for GDPR, ePrivacy Directive, and CCPA compliance. Removal from the Terms homepage suggests either relocation to an alternative delivery mechanism or elimination of the consent interface. Organizations relying on Canva should confirm that cookie consent is still being collected to maintain regulatory compliance.
Added disclosure: Google services use your network connection in the background for updates and operations; you pay associated data costs.
Why it matters: The updated terms establish explicit disclosure that Google services use network connections in the background and assign cost responsibility to users. This clarification formalizes a practice that may have been opaque to users relying on unlimited or variable data plans, and the revised medical disclaimer language provides additional specificity about non-reliance on health-related content.
July 30, 2026
Formatting and navigation structure updated in NotebookLM's Generative AI terms; no substantive policy changes detected.
Why it matters: This change does not materially affect the operational terms governing NotebookLM's Generative AI service. The reorganization is administrative in nature and does not alter user rights, obligations, data collection practices, or AI training permissions.
Extended subscription pre-charge window from 24 to 48 hours; added user liability for uncollected amounts and backup payment method charges.
Why it matters: The updated terms materially increase financial obligations and reduce advance notice for subscription users. The 48-hour pre-charge window and explicit liability for failed-payment retry attempts mean users have less time to manage their payment methods before charges occur. The new explicit liability for background data fees from system services significantly expands what users may owe if their devices are actively updating or communicating without active user interaction, a cost that may be substantial in regions with expensive mobile data. The backup payment method authorization without explicit prior consent may violate consumer protection standards in several jurisdictions, increasing risk of chargebacks and regulatory scrutiny.
Clarified AWS IoT SiteWise Scenario Discovery safety constraints; expanded AI Services definition; reinforced human-oversight requirements for vehicle systems.
Why it matters: The updated terms establish explicit use-case boundaries for AWS IoT SiteWise Scenario Discovery in safety-critical contexts, particularly vehicle systems, and clarify that AWS assumes no liability for uses that violate these boundaries. This change directly affects organizations developing vehicle systems, autonomous features, or regulatory-compliance tools that rely on AWS IoT services; they must now confirm that their architecture includes independent human review and does not treat AWS outputs as autonomous safety validators.
Repository metadata updated; underlying MIT License terms remain unchanged.
Why it matters: This change has no operational significance. The MIT License terms governing DeepSeek-R1 software use, modification, and distribution remain unchanged. The detected update reflects only administrative repository metadata changes.
Product feature renamed from 'Serverless RL' to 'Serverless Training' in platform documentation.
Why it matters: This change reflects a product feature recategorization in Weights & Biases' service documentation. The updated label 'Serverless Training' may more accurately describe the product capability or reflect a rebranding decision, but the underlying service terms and user obligations remain unchanged.
Stay ahead of the changes
You're seeing a fraction of what's changing
ConductAtlas monitors every tracked platform and captures every policy update.
Added product navigation terms to Trust Center marketing tools section.
Why it matters: This change is purely operational and informational. Writer added marketing tool terms to its product navigation menu within the Trust Center. No material operational, compliance, or user-facing implications result from this update.
Added resource category listings (SEO, AEO) in navigation. No substantive policy changes.
Why it matters: This change does not materially affect how privacy practices operate. The update adds navigation labels only and does not modify data handling, consent, disclosure, or user rights.
Added SEO and AEO to resources navigation menu; no contractual changes.
Why it matters: This change does not materially affect how Writer's service operates or what users are obligated to accept. It is a navigation menu enhancement that expands visible resource categories and does not alter binding terms, data practices, or user rights.
Clarified terminology in text messaging privacy protections from 'short code program' to 'text messaging program.'
Why it matters: The updated terminology aligns Intuit's privacy language with industry standard terminology for SMS collection. The substantive protection—that personal information from text signups will not be shared, sold, or rented for third-party marketing—remains intact and unaffected by this change.
Reorganized Model Card Guidelines documentation navigation; no substantive policy change detected.
Why it matters: This change is a documentation navigation restructuring. The substantive guidance on model card creation and governance standards remains unchanged.
Added 'Earn on Uniswap' to privacy policy help documentation index.
Why it matters: This change has no material operational significance. It adds a help documentation topic reference to the privacy policy's table of contents, which does not alter privacy practices, data handling, or user rights.
Added formal Data Protection Officer disclosure and direct privacy contact channels.
Why it matters: The updated policy now provides explicit, accessible contact channels for data protection inquiries, satisfying GDPR and UK GDPR transparency requirements. This clarifies how users and organizations can exercise data subject rights and raises questions or concerns about personal data processing directly with a designated officer.
Reorganized privacy policy help center navigation links; no substantive policy changes detected.
Why it matters: While this change was captured on July 30, 2026, it represents a navigation and documentation restructuring rather than a substantive modification to privacy practices or user protections. No material operational or legal implications are indicated by the change itself.
OpenSea revised its Privacy Policy with a one-sentence modification affecting a numerical value.
Why it matters: OpenSea's Privacy Policy was revised on July 30, 2026. The change involves a one-sentence modification with a numerical adjustment, but without clarity on what the modified sentence addresses or what the numerical value represents, the operational significance for data handling, retention, or user privacy cannot be determined.
Price reference adjusted from $1,922.45 to $1,908.39 in OpenSea Terms of Service.
Why it matters: The modification appears to be a single numerical adjustment in the Terms of Service, detected on July 30, 2026. Without fuller context of the sentence and its application, the practical operational significance remains unclear and should be clarified before assessing compliance or consumer impact.
Expanded browser extension data collection disclosures; added ITIN to personal information types collected.
Why it matters: The updated privacy notice provides more granular disclosure of what merchant page data Affirm's browser extension collects and how it is used. This expanded transparency clarifies the scope of browser extension monitoring and allows users to make more informed decisions about whether to use the extension, with explicit controls now stated (disable, uninstall, hide, or avoid the in-app browser where alternatives exist).
Google restructured Privacy Policy navigation menus; no changes to substantive privacy terms or data practices.
Why it matters: This change does not affect the operational privacy terms users operate under. The substantive policy language remains identical; this is a navigation and structural reorganization of the same content. No changes to Google's data practices, user controls, or privacy protections are introduced by this update.
Navigation restructuring in Google Terms of Service; substantive terms unchanged.
Why it matters: This change does not materially affect the terms users operate under. The substantive policy language, effective date, and legal obligations remain the same; only the presentation and navigation structure have been reorganized.
Updated featured research links in GPT-5.5 System Card; swapped articles on AI safety and robustness for benchmarking and scientific computing research.
Why it matters: The updated system card maintains its core governance and safety framework while curating featured research to highlight recent advances in AI benchmarking and scientific computing. This reflects OpenAI's documentation practices but does not alter the substantive disclosures or operational safeguards that define GPT-5.5 deployment.
Updated related-articles links in GPT-5 System Card; replaced July 15-20 safety articles with July 28-29 AI benchmarking and scientific computing content.
Why it matters: The substantive safety assessment in the system card remains unchanged. This is an editorial update to supplementary resources that does not alter OpenAI's stated precautionary approach or findings regarding biological harm capability thresholds. Organizations relying on system card documentation for governance or due diligence purposes can continue to reference the unchanged core risk assessment.
Updated daily. New changes added as detected.