Automation Anywhere

Structured Summary of Important Changes

1. Overall nature of the diff

The diff appears heavily corrupted by webpage-navigation and marketing content being inserted into the legal text. Large portions of the original Terms are repeated, while headings, paragraphs, and legal language are interspersed with navigation labels, product descriptions, language selectors, and promotional material.

Examples include inserted text such as:

• “Open Login Search Contact Us”
• “AI Without the Hype”
• “Featured Solutions”
• “Get Community Edition”
• “What is Agentic AI?”
• “Automation Anywhere to acquire boost.ai”

These insertions do not appear to form coherent contractual provisions. The revised document should be checked against the actual published Terms before relying on it.

2. Apparent deletion of the opening contractual language

The original opening language stating that the Terms are a “binding agreement between you and Automation Anywhere” appears to be replaced, at least in portions of the diff, by website-navigation and marketing text.

If this reflects the actual revised document, it could create substantial uncertainty about:

• Whether the Terms are clearly presented as a binding agreement;
• The identity of the contracting party;
• The scope of the Sites covered;
• The incorporation of Content and User-Generated Content definitions.

However, the complete original language later appears repeatedly in the diff, suggesting this may be a scraping or rendering error rather than an intentional legal amendment.

3. AI-related changes

No express AI-training authorization identified

The diff does not clearly add language authorizing Automation Anywhere to use customer data, Site data, prompts, inputs, outputs, or User-Generated Content to:

• Train artificial-intelligence models;
• Fine-tune or improve generative-AI systems;
• Develop machine-learning products;
• Use data for model evaluation or benchmarking.

There is no clear new “training,” “machine learning,” “foundation model,” or equivalent provision.

Existing provisions remain relevant to AI use

The existing User-Generated Content clause grants Automation Anywhere and third parties with access a broad, worldwide, transferable, royalty-free, sublicensable license to:

• Copy, distribute, modify, reverse engineer, display, broadcast, record, share, create derivative works of, and otherwise use User-Generated Content.

This language could potentially be broad enough to support use of User-Generated Content in AI-related development, depending on context and applicable privacy law, but it does not expressly mention AI training. The provision also states that submitted suggestions, ideas, comments, and feedback may be used without restriction or compensation.

Users additionally acknowledge that personal information included in User-Generated Content may be disclosed to third parties, with limited responsibility assumed by Automation Anywhere for unauthorized access or misuse.

4. Other continuing risks

• Terms may be changed unilaterally and become effective immediately upon notice, including notice by website posting or email.
• User-Generated Content rights are broad and may survive practical removal or termination.
• Liability is broadly excluded and capped at the lesser of fees paid in the preceding 12 months or $100.
• Users provide broad indemnification.
• California law and Santa Clara County courts apply.
• Claims generally must be brought within one year.

5. Recommended action

Obtain a clean, line-by-line version of the revised Terms. In particular, confirm whether a separate AI, product, privacy, or data-processing policy governs customer inputs and outputs, because this diff does not reliably show a substantive AI-training amendment.

Summary of the Change

• The document states that approximately 11,854 words were removed.
• No replacement language or details of the deleted provisions are provided.
• Because the actual deleted text is unavailable, it is not possible to determine which legal rights, obligations, limitations, or protections were removed.

AI Training and Customer Data

• The supplied diff does not identify whether customer data may be used to train AI models.
• It is therefore impossible to confirm whether the deletion:
• Removed a prohibition on using customer data for AI training;
• Removed customer consent or opt-out rights;
• Expanded or narrowed the provider’s rights to use customer content, prompts, inputs, outputs, or usage data;
• Changed whether data may be used to improve models, analytics, or services;
• Removed confidentiality, de-identification, retention, or deletion requirements; or
• Changed restrictions on sharing data with affiliates, vendors, or AI providers.

Potential Legal Risks

The large-scale deletion could materially affect:

• Data-use rights: Deleted restrictions may permit broader use of customer data.
• AI model training: A deleted prohibition or consent requirement could allow customer data to be used for training or improving models.
• Confidentiality and privacy: Deleted safeguards could reduce protection for confidential, personal, or regulated information.
• Security and retention: Deleted security controls, deletion commitments, or incident-notification obligations could increase risk.
• Liability and remedies: Deleted indemnities, warranties, liability caps, audit rights, or termination rights could shift risk to the customer.
• Compliance: Deleted data-processing terms could affect compliance with privacy and sector-specific laws.

Assessment Limitation

This diff is insufficient for a reliable legal comparison. “Removed approximately 11,854 words” shows the scale of the change but not its substance. The deleted text, or at least the sections addressing data use, AI training, confidentiality, privacy, security, retention, and liability, should be reviewed before accepting the revised document.

Summary

Key Limitation

The supplied diff states only:

> “Added approximately 11,854 words to the document”

It does not include the text of the additions, deletions, or replacements. As a result, it is not possible to identify specific legal changes, including changes to:

• Customer data rights or ownership
• Permitted uses of customer data
• Data sharing or disclosure
• Confidentiality obligations
• Security requirements
• Retention and deletion
• Liability and indemnification
• Intellectual-property rights
• AI training or model-improvement practices

AI Training and Model Use

No conclusions can be drawn about whether the agreement now permits the provider to:

• Use customer data to train, fine-tune, or improve AI models
• Use customer prompts, inputs, outputs, files, or personal information for training
• Use data for product development, benchmarking, analytics, or abuse monitoring
• Combine customer data with other customers’ data
• Create or retain de-identified, aggregated, or derived data
• Allow human reviewers or service providers to access data
• Opt customers in or out of AI-training programs
• Apply different data-use rules to consumer, business, or enterprise accounts

These provisions should be specifically reviewed in the added text.

Potential Risk from the Size of the Addition

An addition of approximately 11,854 words is potentially material. Even without the text, the size of the change suggests that the agreement may contain substantial new terms, including provisions that could alter the customer’s obligations, remedies, privacy rights, or consent to data use. The absence of the actual language prevents assessment of whether the changes are favorable, neutral, or adverse.

Information Needed

Please provide the complete redlined text or the actual additions, deletions, and replacements. The AI-related sections should receive particular attention, including terms such as:

• “train,” “training,” “fine-tune,” or “improve models”
• “inputs,” “outputs,” “customer data,” or “content”
• “de-identified” or “aggregated data”
• “service improvement” or “product development”
• “retain,” “share,” “disclose,” or “第三-party providers”
• “opt out,” “consent,” or “enterprise data”

Structured Summary of Important Changes

1. Website terms replace general marketing/navigation content

The revised material replaces website navigation, promotional text, and product descriptions with formal Website Terms. The Terms state that they are a binding agreement between the user and Automation Anywhere, Inc. and apply to:

• www.automationanywhere.com;
• subdomains and other Automation Anywhere websites;
• website content; and
• user-generated content.

Risk: The Terms apply broadly to anyone accessing or using the Sites, including automated users, and acceptance occurs simply by accessing or using the Sites.

2. Broad user-generated-content license

Users who upload, share, or otherwise provide content grant Automation Anywhere and third parties to whom it provides access a:

• transferable;
• royalty-free;
• sublicensable;
• worldwide;
• non-exclusive license.

The license permits copying, distributing, modifying, reverse engineering, displaying, broadcasting, recording, sharing, creating derivative works, and otherwise using the content.

Users also acknowledge that personal information included in user-generated content may be disclosed to third parties, and Automation Anywhere disclaims responsibility for unauthorized access or misuse.

Risk: This is a very broad license with no express limitation on purpose, duration, retention, confidentiality, or deletion. Customers should not submit confidential business information, personal data, regulated data, trade secrets, or content belonging to third parties unless separately authorized.

3. AI-model training and data use

The diff does not expressly state that customer or user data will be used to train, fine-tune, evaluate, or improve AI models.

However, the broad user-generated-content license allows Automation Anywhere to “otherwise use” submitted content and to create derivative works. It also permits access by third parties. Depending on interpretation, this language could potentially support data processing for analytics, product improvement, or AI-related purposes, although it does not specifically authorize AI training.

The Terms separately state that feedback, suggestions, ideas, and comments may be used without restriction or compensation.

Key risk: There is no explicit “no training” commitment, prohibition on training, data-isolation promise, de-identification requirement, or limitation preventing disclosure of submitted content to AI providers or other subcontractors. The incorporated Privacy Policy may contain additional relevant terms and should be reviewed.

4. Unilateral changes

Automation Anywhere may revise the Terms or impose new conditions at any time. Changes may be communicated by posting them on the Sites or by email and become effective immediately upon notice. Continued use constitutes acceptance.

Risk: Users may become bound by materially different terms without signing a new agreement or receiving individualized notice.

5. Commercial-use limitations and separate agreements

The Sites and Content are generally limited to personal, non-commercial use. Commercial downloads or software may be governed by separate Automation Anywhere license agreements, which control in case of conflict.

Risk: Business users may inadvertently breach the Terms by using website materials commercially or assuming that these Terms govern related products and services.

6. Liability and dispute provisions

The Terms include:

• “as-is” and “as-available” disclaimers;
• broad exclusions of damages;
• an aggregate liability cap of the lesser of fees paid in the preceding 12 months or $100;
• a broad user indemnity;
• California law and Santa Clara County courts; and
• a one-year deadline to bring claims.

Risk: Remedies are substantially limited, particularly for data loss, service interruptions, misuse of uploaded content, or privacy-related harm.

Diff Analysis

Overall Change

The diff indicates that approximately 11,854 words were added, but it does not include the text of those additions. Because the substantive language is missing, it is not possible to reliably identify:

• Changes to rights or obligations
• New fees, limitations, warranties, or indemnities
• Changes to termination, governing law, or dispute provisions
• Privacy, confidentiality, or security risks
• Any provisions concerning use of customer data to train AI models

AI Training and Customer Data

No conclusion can be reached from the supplied diff. The placeholder does not reveal whether the added terms:

• Permit the provider to use customer data, prompts, inputs, outputs, or usage data to train or improve AI models
• Permit use of data for generalized model training, product development, analytics, or benchmarking
• Distinguish between personal data, confidential information, and de-identified or aggregated data
• Provide an opt-out or require affirmative consent
• Apply training rights to subcontractors, affiliates, or third-party model providers
• Allow the provider to retain data after termination
• Allocate responsibility for intellectual-property infringement arising from model training or outputs
• Impose restrictions on submitting regulated, confidential, or personal information

These issues should be treated as unresolved and requiring review.

Risk Assessment

The addition of 11,854 words is potentially significant, but the risk level cannot be assessed without the actual language. A large new section could materially alter the agreement, including by introducing incorporated policies, broad licenses, data-use rights, or limitations of liability.

Information Needed

Please provide the actual added text, or a complete redline showing the additions and deletions. At minimum, the review should include provisions titled or addressing:

• Data use and ownership
• Privacy and security
• Artificial intelligence or machine learning
• Service improvement or product development
• Confidentiality
• Intellectual property
• Retention and deletion
• Subprocessors or third-party providers
• Warranties, indemnities, and liability limits

Summary of Important Changes

1. New binding Website Terms

• The page changes from general marketing/navigation content to a formal agreement between the user and Automation Anywhere, Inc.
• The Terms cover www.automationanywhere.com, subdomains, and other Automation Anywhere websites, collectively defined as the “Sites.”
• The Terms apply to use of Site content and User-Generated Content.
• Accessing or using the Sites—whether manually or automatically—is deemed acceptance.
• Users must be at least 18 or the age of majority where they reside.

Risk: Users may become contractually bound merely by browsing or using the Sites, without signing an agreement.

2. Broad unilateral amendment rights

• Automation Anywhere may revise the Terms or impose new conditions at any time.
• Changes may be communicated by posting them on the Sites or sending an email.
• Continued use after notice constitutes acceptance.

Risk: Terms can change without an affirmative signature or click-through acceptance, potentially creating uncertainty about which version applies.

3. Expanded restrictions on use

The Terms impose broad restrictions, including prohibitions on:

• Automated access or scraping;
• Mass collection of Site information;
• Unauthorized access to accounts or systems;
• Disruption or overloading of the Sites;
• Copying, storing, modifying, redistributing, or commercially using Content without permission.

Commercially used downloaded materials may also be subject to separate Automation Anywhere license agreements, which control in the event of conflict.

4. User-Generated Content license

Users grant Automation Anywhere and third parties it authorizes a:

• Transferable;
• Royalty-free;
• Worldwide;
• Sublicensable;
• Non-exclusive

license to copy, distribute, modify, reverse engineer, display, broadcast, record, share, create derivative works from, and otherwise use User-Generated Content.

Users remain responsible for their submissions and must warrant that they have all necessary rights, including rights relating to privacy, publicity, confidential information, and intellectual property.

Risk: The license is extremely broad and has no express limitation by purpose, duration, retention period, or deletion obligation. It could permit extensive commercial exploitation or disclosure of submitted content.

5. AI and model-training implications

• The revised content introduces references to Automation Anywhere’s AI offerings and “AI Terms,” including AI Agent Studio, Automation Co-Pilot, and agentic automation products.
• However, the supplied diff does not clearly add an express provision stating that customer data will be used to train, fine-tune, or improve AI models.
• The broad User-Generated Content license includes “modify,” “reverse engineer,” “create derivative works,” and “otherwise use,” which could arguably encompass use in AI development or training if submitted material qualifies as User-Generated Content.
• The Terms do not expressly exclude confidential information, personal data, customer content, prompts, outputs, or submissions from model training. They also do not specify whether data is de-identified, whether third-party model providers receive it, or whether users can opt out.

Key risk: There is no clear training authorization or prohibition. Customers should seek written clarification and confirm that customer data, confidential information, prompts, and outputs will not be used for model training without express consent.

6. Liability and dispute provisions

• Broad “as is” disclaimers apply.
• Liability is capped at fees paid in the prior 12 months or $100, whichever is less.
• Consequential, indirect, punitive, and data-loss damages are excluded.
• Users must indemnify Automation Anywhere for claims arising from their use or violations.
• California law and Santa Clara County courts apply.
• Claims must be brought within one year.

Overall assessment: The new Terms materially strengthen Automation Anywhere’s control over Site use and submitted content, while providing limited customer remedies. The AI/data-use position remains ambiguous and should be addressed contractually.

Summary

The diff does not include the substance of the new language. It only states:

> “Added approximately 11,847 words to the document”

Accordingly, it is not possible to identify the legal effect of the changes, including whether the customer’s data may be used to train AI models.

AI Training and Data Use

The provided diff contains no specific terms addressing:

• Whether customer data may be used to train, fine-tune, validate, or improve AI models
• Whether customer prompts, inputs, outputs, files, or usage metadata are retained
• Whether data is used for product development, analytics, or service improvement
• Whether customer data is anonymized, aggregated, or de-identified before use
• Whether data is shared with affiliates, vendors, subprocessors, or model providers
• Whether customers can opt out of AI training or request deletion
• Whether the provider claims ownership or broad licensing rights in customer data or outputs
• Whether confidential information or personal data may be included in training datasets
• Security, confidentiality, retention, or cross-border transfer protections applicable to such use

Other Legal Risks

Because the actual additions are missing, the following issues also cannot be evaluated:

• Changes to liability caps, indemnities, warranties, or disclaimers
• Expanded provider rights to suspend or terminate services
• Changes to fees, renewals, or payment obligations
• New audit, compliance, or regulatory obligations
• Changes to intellectual-property ownership or licenses
• Privacy, data-processing, security, or breach-notification provisions
• Governing law, venue, arbitration, or dispute-resolution terms
• Unilateral amendment rights or incorporation of external policies

Information Needed

To perform a meaningful review, provide the actual added, deleted, and replacement language—not merely the word-count description. The AI-related provisions should be reviewed especially carefully for broad rights to use customer data, absence of an opt-out, indefinite retention, use by third-party model providers, and rights to use data containing confidential or personal information.

Summary of Changes

Scope of the Diff

• The diff indicates that approximately 11,847 words were removed from the document.
• No replacement language or additions are shown.
• Because the deleted text is not provided, the specific legal effects cannot be determined.

Key Legal Risks

Loss of Customer Protections

The removed provisions may have contained important protections concerning:

• Confidentiality and data security
• Limits on the provider’s use of customer data
• Data ownership and intellectual-property rights
• Service commitments and remedies
• Indemnification and liability limits
• Audit, compliance, or regulatory obligations
• Termination and data-deletion requirements

A wholesale deletion creates a material risk that customer rights, restrictions, or remedies have been eliminated.

Ambiguity and Interpretation Risk

If the deleted provisions were not replaced, the remaining agreement may:

• Leave important subjects unaddressed
• Shift risk to the customer by default
• Create conflicts with other sections
• Make enforcement more uncertain
• Remove definitions or cross-references needed to interpret surviving clauses

AI Training and Data Use

• The provided diff does not identify whether customer data may be used to train, fine-tune, evaluate, or improve AI models.
• It is therefore impossible to determine whether:
• A prohibition on AI training was removed
• Customer data may now be used for model development
• Human review or data sharing with affiliates or vendors is permitted
• Customer prompts, outputs, files, or usage metadata may be retained
• The provider must delete data or exclude it from training upon request
• Separate consent is required for AI-related data use

The deletion of approximately 11,847 words could have removed material AI-data restrictions, but that cannot be confirmed from the information supplied.

Recommended Review

Obtain:

1. The full pre-change and post-change agreements;

2. The actual deleted text, including section headings; and

3. Any updated privacy, data-processing, or AI-use policies incorporated by reference.

Until reviewed, treat the change as potentially material and avoid assuming that customer data remains protected from AI training or other secondary uses.

Structured Summary of Important Changes

1. Major document-integrity problem

The diff appears to replace substantial portions of the legal Terms with website navigation, marketing copy, product listings, contact prompts, language selectors, and other scraped page content. For example, the opening agreement language is replaced by text such as:

• “Open Login Search Contact Us”
• “Have a question?”
• “AI Without the Hype”
• Product, industry, and resource menus

Although much of the original legal text later appears again, the resulting document is highly duplicated and interspersed with non-legal content. This creates significant risks:

• Uncertainty about which text forms the operative agreement
• Possible ambiguity or lack of mutual assent
• Broken section numbering and cross-references
• Difficulty proving what users saw or accepted
• Potential website publication and consumer-protection concerns

The document should be reconstructed from a clean source before publication.

2. No express AI-training authorization identified

The diff does not clearly add or remove an express clause stating that customer data, prompts, inputs, outputs, or other user data may be used to train AI models.

References to AI—such as “AI Agent Studio,” “Agentic AI,” “AI Governance,” and marketing language—appear to be website content rather than contractual provisions. They do not themselves establish a clear data-training right.

Accordingly, the Terms remain unclear on:

• Whether submitted data may be used to train, fine-tune, or improve AI models
• Whether data may be used for product development or benchmarking
• Whether data is de-identified before such use
• Whether customer data is shared with model providers or other third parties
• Whether customers can opt out
• Whether inputs, outputs, or confidential information are excluded from training

A separate privacy policy, product agreement, or data-processing addendum may address these issues, but the diff does not show that.

3. Existing user-content rights remain broad

The reproduced User-Generated Content clause grants Automation Anywhere and third parties with access a transferable, worldwide, royalty-free, sublicensable license to:

• Copy, distribute, modify, reverse engineer, display, broadcast, record, share, and create derivative works
• Otherwise use the User-Generated Content

This is broad enough that a user could argue it permits data-mining or AI-training activities, even though training is not expressly mentioned. It also permits third-party access and provides no clear confidentiality, deletion, retention, or opt-out protections.

The clause separately states that feedback may be used “without any restriction or compensation.” This could cover suggestions or comments about the Sites and potentially permit their use in model or product development.

4. Customer and personal-data risks

Users acknowledge that personal information included in User-Generated Content may be disclosed to third parties, and Automation Anywhere disclaims responsibility for unauthorized access or misuse. This places substantial risk on users and may be inadequate for confidential business data or regulated personal information.

5. Other continuing risk allocations

The Terms continue to include:

• Broad “as-is” disclaimers
• Extensive exclusions of consequential and other damages
• Liability capped at fees paid or $100, whichever is less
• Broad customer indemnification obligations
• California governing law and Santa Clara County venue
• One-year limitation period for claims
• Immediate effectiveness of changes upon notice, with continued use treated as acceptance

These provisions should be reviewed against applicable consumer, privacy, and data-protection laws.

Summary of Changes

Overall Change

• Approximately 11,847 words have been removed from the document.
• Because the actual deleted text is not provided, it is not possible to determine which contractual rights, obligations, limitations, or protections were removed.

Customer Data and AI Training

• The diff does not identify whether any provisions concerning customer data, artificial intelligence, machine learning, or model training were deleted.
• In particular, it is not possible to confirm whether the removed language addressed:
• Use of customer data to train, fine-tune, or improve AI models;
• Whether customer data may be used for generalized or product-wide model training;
• De-identification, aggregation, or anonymization requirements;
• Customer consent or opt-out rights;
• Restrictions on using customer prompts, inputs, outputs, or personal information;
• Data retention and deletion after use;
• Human review or automated processing of customer data;
• Ownership of, or rights in, models trained using customer data; or
• Security and confidentiality obligations applicable to AI systems.

Potential Legal Risks

The deletion could potentially remove important customer protections, including:

• Confidentiality protections, if restrictions on using or disclosing customer information were deleted.
• Purpose limitations, if language limiting data use to providing the services was removed.
• Consent or opt-out mechanisms, particularly for AI training or product improvement.
• Data protection commitments, such as deletion, retention, security, or compliance requirements.
• Intellectual property protections, including ownership or licensing terms relating to customer data and AI-generated outputs.
• Liability limitations or indemnities, which may affect responsibility for misuse, privacy violations, or security incidents.
• Audit, transparency, or notice rights, potentially reducing the customer’s ability to understand or control data use.

Assessment Limitation

This change cannot be reliably analyzed without the text of the deleted provisions or a prior version of the document. The deletion should be reviewed specifically for any clauses containing terms such as “customer data,” “content,” “inputs,” “outputs,” “train,” “train models,” “machine learning,” “AI,” “improve,” “develop,” “de-identify,” “aggregate,” “retain,” or “service improvement.”

Structured Summary of Important Changes

1. Diff integrity and apparent document replacement

• The diff is highly corrupted or appears to compare the Terms page against large amounts of website navigation, marketing, and footer content.
• Large portions of the legal Terms appear in deletion blocks and are then reproduced later as additions or unchanged text.
• As a result, it is not possible to reliably determine whether the operative legal Terms were substantively amended, or whether the changes primarily reflect a website redesign, localization, or scraping error.
• The final displayed content appears to include the existing Website Terms substantially intact, together with extensive new website content.

2. New AI-related website content

The additions include references to:

• “Automation Anywhere AI”
• “AI in Action”
• “AI Technology”
• “AI Agent Studio”
• “Automation Co-Pilot”
• “Agentic process automation”
• “AI Governance”
• “AI Agents”
• “Document Automation”
• “Responsible AI”

These appear to be marketing, navigation, product, or educational content rather than contractual provisions. Their inclusion may broaden the apparent scope of the “Sites,” but the diff does not clearly add operative terms governing those products.

3. Customer data and AI-model training

No express new training authorization identified

The diff does not clearly add language stating that customer data, prompts, inputs, outputs, User-Generated Content, or other submitted information may be used to train, fine-tune, evaluate, or improve AI models.

There is also no clear addition addressing:

• Whether data is used for model training by default;
• Whether training is conducted by Automation Anywhere or third-party providers;
• Whether data is anonymized or aggregated first;
• Customer opt-out or deletion rights;
• Ownership of model outputs or derivatives;
• Confidentiality protections for data used in AI development.

Existing provisions remain potentially relevant

The reproduced Terms contain a broad license for User-Generated Content, allowing Automation Anywhere and third parties it authorizes to:

> copy, distribute, modify, reverse engineer, display, broadcast, record, share, create derivative works of, and otherwise use

that content. The license is transferable, royalty-free, sublicensable, worldwide, and non-exclusive.

Although this does not expressly mention AI training, its broad wording could arguably encompass using submitted content for model development, depending on the nature of the content and applicable law. Users also acknowledge that personal information in User-Generated Content may be disclosed to third parties.

Submitted suggestions, ideas, comments, and feedback may be used “without any restriction or compensation,” which may create additional risk if feedback contains confidential or commercially sensitive material.

4. Other continuing risks

• Terms may be changed immediately upon notice, including by posting on the Sites or sending email; continued use constitutes acceptance.
• Automation Anywhere disclaims responsibility for unauthorized access to or misuse of personal information in User-Generated Content.
• Liability is broadly excluded and capped at fees paid in the prior 12 months or $100, whichever is less.
• Users provide broad indemnification for claims arising from site or content use.
• The Privacy Policy is incorporated by reference, so data-use terms may appear there rather than in these Terms.

Bottom line

The diff does not show a clear contractual change expressly authorizing AI-model training on customer data. However, the broad existing User-Generated Content license and incorporated Privacy Policy should be reviewed carefully before submitting confidential, personal, or proprietary information.

Summary of Important Changes

Scope of the Change

• Approximately 11,847 words were removed from the document.
• No replacement language or surviving text was provided.
• Because the deleted provisions are not shown, the specific legal and commercial effects cannot be determined from this diff alone.

Customer Data and AI Training

• The diff does not identify whether customer data may now be used to train AI models.
• However, deleting a substantial portion of the agreement could have removed provisions addressing:
• Whether customer data is used to train, fine-tune, test, or improve AI models;
• Whether customer data is de-identified, aggregated, or anonymized before such use;
• Whether the provider must obtain customer consent before using data for AI training;
• Whether customer content is excluded from model training by default;
• Ownership and licensing rights in customer data, inputs, outputs, and derived data;
• Restrictions on using confidential, personal, or regulated information for AI purposes;
• Data-retention, deletion, and opt-out rights;
• Security obligations and subprocessors involved in AI processing.

Potential Risks

• Material uncertainty: The deletion may remove important customer protections, but the available diff does not establish which ones.
• Expanded data-use risk: If deleted restrictions previously limited secondary use, the customer could have fewer contractual protections against using its data to develop or improve AI systems.
• Reduced transparency: Deleted disclosure, consent, or opt-out provisions could make it harder to determine how customer data is processed.
• Confidentiality and privacy exposure: Deleted limitations could permit broader use of confidential information or personal data, potentially creating regulatory, confidentiality, or contractual concerns.
• Ownership and licensing ambiguity: Removing data-rights language could create uncertainty over whether the provider receives a license to use customer data, outputs, or derivatives.
• Remedy risk: Deleted audit rights, indemnities, liability provisions, or breach-notification requirements could weaken the customer’s remedies if data is misused.

Recommended Review

Obtain the full pre-change and post-change agreements, or at least the text of the deleted provisions. Priority should be given to sections titled Data Use, Artificial Intelligence, Machine Learning, Confidentiality, Privacy, Security, Content, Intellectual Property, Retention/Deletion, and Service Providers. Until that comparison is available, the impact of the deletion—particularly on AI-model training—should be treated as unresolved and potentially material.

Summary of Important Changes

1. Apparent replacement of legal text with website navigation/content

The diff shows substantial portions of the Terms replaced or interspersed with website navigation, marketing copy, product listings, event information, language selectors, and footer content. Examples include:

• “Open Login Search Contact Us Talk to Sales”
• Product and solution menus
• “AI Without the Hype”
• “Get Community Edition”
• Gartner marketing claims
• Contact and support information

These changes appear more consistent with a website scraping, template, or rendering error than deliberate legal amendments. If this is intended to be the operative Terms page, it creates serious uncertainty about:

• Where the legal terms begin and end;
• Whether the agreement remains readable and enforceable;
• Which content is contractual versus merely promotional; and
• Whether definitions and section references remain accurate.

The document also appears duplicated multiple times, increasing ambiguity about the operative version.

2. No express AI-training authorization identified

The diff does **not clearly add a clause expressly stating that customer data, User-Generated Content, prompts, inputs, outputs, or other information may be used to train, fine-tune, improve, or evaluate AI models.

Accordingly, there is no clear new contractual statement that Automation Anywhere may train AI models on customer data.

However, the existing User-Generated Content license remains important. It grants Automation Anywhere and third parties with access a broad, worldwide, royalty-free, transferable, sublicensable license to:

• Copy, distribute, modify, reverse engineer, display, broadcast, record, share;
• Create derivative works; and
• “Otherwise use” User-Generated Content.

That language is broad enough to create an argument that certain uses related to AI development or model improvement could be permitted, particularly for content voluntarily submitted through community, support, or other interactive features. It does not contain limitations such as “solely to provide the Sites,” “no model training,” deletion obligations, or de-identification requirements.

3. Data-disclosure and privacy risk

The Terms continue to state that User-Generated Content containing personal information may be disclosed to third parties, and that Automation Anywhere is not responsible for unauthorized access to or misuse of that information. This creates a significant risk for users who upload confidential, personal, regulated, or customer data.

The Terms refer generally to the Privacy Policy but do not, in this diff, provide AI-specific data-use restrictions or guarantees. The Privacy Policy and any product-specific agreement should therefore be reviewed for training, retention, subprocessors, and data-isolation provisions.

4. Other continuing risks

The Terms retain broad provisions that:

• Permit immediate modification of the Terms upon notice;
• Allow suspension or termination at any time;
• Limit liability to the lesser of fees paid in the prior 12 months or $100;
• Exclude consequential and data-loss damages; and
• Impose broad indemnification obligations on the user.

Recommended action: Confirm the intended clean legal text, investigate the apparent rendering/scraping corruption, and obtain express written clarification that customer data and confidential User-Generated Content will not be used to train AI models unless separately authorized.