Prior version of these AI Service Terms available here.

*THE FAQs BELOW ARE NOT INCORPORATED INTO BENCHLING’S OFFICIAL AI SERVICE TERMS and ARE FOR INFORMATIONAL PURPOSES ONLY*

Frequently Asked Questions

A Reviewer’s Guide to the Benchling AI Service Terms

This guide is intended for our customers’ legal and procurement teams. It provides context on Benchling’s AI Services to streamline your review of our AI Service Terms and explain how they are structured alongside the Benchling Main Services Agreement.

WHAT ARE BENCHLING’S AI SERVICES AND HOW DO THESE AI TERMS RELATE TO OUR EXISTING AGREEMENT?

Benchling’s AI Services are the artificial intelligence and machine learning features and functionality built into the Benchling platform. These include AI-powered tools designed specifically for bioscience R&D – such as intelligent agents, predictive and generative scientific models, and workflow automation – all aimed at helping your teams work faster and uncover deeper insights from your data.

The AI Service Terms function as a focused addendum to your existing Services Agreement (whether that’s our MSA, Terms of Service, or another written agreement). They do not replace your existing agreement; they supplement it. In the event of a conflict, the AI Service Terms govern only with respect to AI Services and certain definitional updates. All other terms – including dispute resolution, general data protection, and commercial terms – continue to be governed by your Services Agreement.

A current list of available AI Services is always available at https://www.benchling.com/ai-services.

DOES BENCHLING USE MY DATA TO TRAIN AI MODELS?

Benchling will not use your Customer Data – including your AI Inputs and AI Outputs – to train any AI or machine learning models without your express permission or instruction, except where you choose to use an AI Service that is specifically designed to train a model tailored for your use (e.g., a custom model trained on your proprietary datasets). Even in the case where the model is tailored to your use, that model is used only for you and will never be used for any other customer.

We also contractually prohibit our third-party service providers whose services power our AI Services from training on Customer Data.

Benchling may access and use AI Inputs and AI Outputs for a narrow set of operational purposes: to operate, evaluate, improve, and support the AI Services, comply with applicable law, enforce our terms, and keep the platform safe and secure. This is analogous to how any SaaS provider may need to access data to operate, maintain, and improve its services, and it does not include training models on your data.

WHO OWNS THE AI OUTPUTS GENERATED FROM MY DATA?

You do. You retain full ownership of your AI Inputs, and to the extent permitted by applicable law, you own the AI Outputs generated in response to your AI Inputs. Benchling also assigns to you any rights it may have in the AI Output.

This ownership framework is consistent with the approach in our MSA: intellectual property you develop from your Customer Data using the Benchling Services belongs to you.

There is one important nuance to be aware of: because AI systems are trained on broad datasets, AI Output may sometimes reflect underlying patterns or content from training data. This means we cannot guarantee that every AI Output is entirely unique – it is possible that similar or identical outputs could be independently generated for other users who provide similar inputs. This is an inherent characteristic of AI technology, not a limitation of Benchling’s AI Services.

WHAT DOES BENCHLING DO WITH FEEDBACK I PROVIDE ON AI OUTPUTS?

Providing Feedback is entirely voluntary. However, when you do provide it – for example, by using a thumbs up/down rating on an AI Output – you grant Benchling the right to use that Feedback to evaluate and improve the AI Services.

Feedback is distinct from Customer Data. It includes structured signals like ratings and unstructured comments about the AI Services’ performance. Benchling may use Feedback freely to improve the platform for all customers, which benefits the entire Benchling community.

WHERE DOES MY DATA GO WHEN I USE THE AI SERVICES? DOES BENCHLING SHARE IT WITH THIRD PARTIES?

To deliver AI functionality, certain AI Inputs and Customer Data may be processed by Benchling’s AI systems or by third-party AI providers that power elements of the AI Services. Importantly, as noted above, Benchling’s third-party service providers are not permitted to use your Customer Data to train AI models

Benchling maintains the same enterprise-grade security practices for AI Services that apply across the entire platform. Details on our general security practices and certifications are available at our Benchling Security Center, and you can read more about data protection and security for AI at Benchling in our Help Center and our AI Security Whitepaper.

WHY ARE AI OUTPUTS PROVIDED “AS IS” WITHOUT WARRANTIES?

AI and machine learning technology is powerful but inherently probabilistic. Even the most advanced models can produce outputs that are inaccurate, incomplete, or unexpected. This is true across the industry – not unique to Benchling.

Because of this, our AI Services Terms are transparent about the current state of the technology: AI Outputs and Actions are provided “as is,” without warranties of accuracy, completeness, or reliability. This approach is industry-standard for AI services and reflects an honest assessment of the technology’s capabilities today.

You should apply human review and independent verification before relying on AI Outputs – especially for critical decisions. This is good scientific practice, and our terms are designed to support it by making the expectations around AI Output clear and explicit.

WHY DOESN’T BENCHLING’S IP INDEMNITY COVER AI OUTPUTS?

Our MSA includes an industry-standard intellectual property infringement indemnity for the Benchling Services. However, that indemnity does not extend to the AI Services and AI Outputs.

This is because AI Outputs are generated dynamically based on your unique inputs and the underlying model’s training data. Unlike the Benchling platform itself – which is proprietary software under Benchling’s control – AI Outputs are shaped by an enormous range of variables that make it impractical to provide the same IP guarantee. This exclusion is consistent with industry practice.

Benchling stands behind its platform and its proprietary technology. The IP indemnity in your Services Agreement continues to apply to all Benchling Services other than the AI Services.

WHY DO CUSTOMERS INDEMNIFY BENCHLING FOR AI-RELATED CLAIMS?

Benchling’s AI Services are data-agnostic and customer-controlled. You control what data you input, what outputs you use, and what actions you direct the AI Services to take. Because Benchling does not control these decisions, the AI Service Terms include a customer indemnity for third-party claims arising from your use of the AI Services, your AI Inputs, AI Outputs, or Actions.

This is a natural extension of the platform model described in our MSA: because the customer controls the data and how the platform is used, the customer is best positioned to manage the risks associated with that use. The indemnity ensures a balanced risk allocation that reflects this reality.

WHAT ARE THE ACCEPTABLE USE RESTRICTIONS, AND WHY DO THEY EXIST?

The acceptable use rules in Section 3 are designed to protect you, other customers, and the integrity of the Benchling platform. They cover common-sense requirements: use the AI Services lawfully, don’t violate others’ rights, review AI Outputs before relying on them, don’t reverse-engineer the models or our AI Services, and don’t attempt to compromise system integrity.

These rules are standard for enterprise AI services and exist to ensure a safe, reliable, and high-quality experience for all customers. They also protect you by setting clear expectations for responsible use.

CAN BENCHLING UPDATE THESE AI TERMS?

Yes. AI technology and the regulatory landscape around it are evolving rapidly. To ensure our terms remain current with the technology and applicable law, Benchling reserves the right to update the AI Service Terms. We will notify you of any material changes by posting the updated terms on our website or through other reasonable means.

Your continued use of the AI Services after the effective date of any revised terms constitutes your acceptance. This approach allows Benchling to move quickly to incorporate new protections, reflect regulatory changes, and improve the terms over time for all customers.

WHAT ABOUT THIRD-PARTY MODELS AND THIRD-PARTY OFFERINGS?

Benchling may make Third-Party Models and Third-Party Offerings (such as open-source models, third-party datasets, or partner applications) available for use with the AI Services. In some cases, these may be governed by separate third-party terms – not by the AI Service Terms or your Services Agreement.

Benchling does not warrant, support, or assume liability for Third-Party Offerings or how those third parties handle Customer Data. We may make the applicable third-party agreements available for your review, but the relationship with the third-party provider is between you and them.

This approach provides you with flexibility to leverage a rich ecosystem of AI tools within Benchling, while maintaining clear lines of accountability.

HOW DOES BENCHLING HANDLE CUSTOMER MODELS?

A Customer Model is an AI or machine learning model that you own or license and choose to upload, fine-tune, or deploy within the AI Services. Your Customer Models remain your property.

Benchling’s disclaimers of liability apply to your use of Customer Models with the AI Services, consistent with the data-agnostic, customer-controlled nature of the platform: you are responsible for the models you bring to the platform, just as you are responsible for the data you upload.

*THE FAQs ABOVE ARE NOT INCORPORATED INTO BENCHLING’S OFFICIAL AI SERVICE TERMS and ARE FOR INFORMATIONAL PURPOSES ONLY*

These AI Service Terms (“AI Terms”) govern the use of any AI Services (defined below) provided as part of the Benchling Services (also defined below).

The AI Terms are a binding legal agreement between Benchling, Inc. and its affiliates (“Benchling”) and the Customer or user agreeing to these terms (“you”). By enabling or using an AI Service (including by opting in or prompting an AI feature), you agree to be bound by these AI Terms. If you do not agree, you may not use the AI Services. Use of AI Services constitutes acceptance of these terms, which will be deemed agreed to each time you utilize an AI Service. If you are agreeing to these terms on behalf of a Customer (such as your employer) or other legal entity, you represent and warrant that you have the authority to bind that company or other legal entity to these terms. In that case, “you” refers to that Customer or other legal entity.

These AI Terms are an addendum to, and are incorporated by reference into, Benchling’s Terms of Service, Main Services Agreement and, if you are not using Benchling under either of these agreements, the written agreement with Benchling that governs your use of the Benchling Services (collectively these shall be referred to as your “Services Agreement”). In the event of a conflict between these AI Terms and your Services Agreement, these AI Terms shall prevail solely with respect to your use of AI Services. Notwithstanding the foregoing, if any definitions in these AI Terms conflict with definitions in your Services Agreement, then the definitions in these AI Terms shall supersede and replace the equivalent and conflicting definitions in such agreement. Capitalized terms used herein, but not defined either when they first appear or in Section 1 (Definitions) below, shall have the same meaning set forth in your Services Agreement. For the avoidance of doubt, the dispute resolution procedures your Services Agreement shall apply to any dispute arising out of or related to these AI Terms.

1. Definitions

“Acceptable Use Policy” means the policy located at https://www.benchling.com/acceptable-use-policy, which describes rules that apply to any party using any products and services provided by Benchling or any of its affiliates, as well as to any user of any software application or service made available by you that interfaces with the Benchling Services.

“Account Data” means data that relates to Benchling’s relationship with you, including but not limited to your name, contact information, username, user ID, password, responses to security questions, and the name of the organization you are affiliated with. Account Data also includes any data Benchling may need to collect for the purpose of managing its relationship with you, including information used to authenticate users, administer accounts, and provide secure access to Benchling’s platform; identity verification; account settings and configuration; or as otherwise required by applicable laws and regulations. For clarity, Account Data does not include Customer Data.

“Actions” means actions the AI Services are enabled to take on your behalf, including but not limited to software manipulation, data processing, and system interactions.

“AI Input” means Customer Data submitted or made available to an AI Service by you. For clarity, input into an AI Service may consist of both Customer Data and data that is not Customer Data (for example, suggested prompts). Only the Customer Data component of input into an AI Service shall be considered AI Input.

“AI Output” means any Output generated and returned to a Customer or user by an AI Service based on AI Input.

“AI Service(s)”means features and functionality made available as part of the Benchling Services that are labeled, identified, or described by Benchling as including, using, powered by, or being an artificial intelligence (“AI”) or machine learning feature or service. A list of Benchling’s currently available AI Services, including Trials and Betas, can be found here: https://www.benchling.com/ai-services.

“Application Services” means the proprietary cloud services and related offerings hosted on servers under the control or direction of Benchling, as modified from time to time.

“Benchling Services” means the Application Services, Benchling Installed Software, and the Documentation. The Benchling Services do not include Customer Third-Party Services.

“Benchling Installed Software” means Benchling’s proprietary software components or tools made available for installation on your computer systems or devices, including but not limited to for the purpose of integrating the Application Services with Customer Third-Party Services.

"Customer Data" means any data, information, content, or code that: (i) you submit, post, upload, transmit, or otherwise make available to the Application Services, including via Customer Third-Party Services, except for such data specifically excluded in this definition; or (ii) is Output. Customer Data specifically excludes: (1) Benchling Intellectual Property, including but not limited to templates, suggested inputs or prompts, and proprietary datasets, models, or training corpora; (2) Account Data; (3) Usage Data, including any such data that incidentally incorporates data made available to the Application Services by Customer; (4) Feedback, whether structured (e.g., "thumbs up/down" signals, ratings) or unstructured (e.g., free-text submissions); and (5) aggregated and de-identified data.

“Customer Third-Party Services” means any product, add-on, services, or platform not provided by Benchling that Customer uses with the Benchling Services.

“Customer Model” means an AI or machine learning model owned or licensed by you that you upload, fine tune, or deploy in the AI Services.

“Feedback” means comments, suggestions, ideas, enhancement or improvement requests, recommendations, or other feedback that you provide to Benchling related to the Benchling Services.

"Output" means the content, reports, models, visualizations, calculations, files, designs, predictions, code, sequences, and similar information generated directly from Customer Data by you, or by Benchling via the Benchling Services, based on your use or input. For clarity, Output excludes the Benchling Intellectual Property, including where such Benchling Intellectual Property is incorporated into generated outputs (e.g., templates of reports and models).

“Third-Party Model” means an AI or machine learning model owned or licensed by a third party that Benchling uploads, fine tunes, or deploys in the AI Services on your direction and behalf, or that is otherwise made available by Benchling for your use directly with the AI Services. For clarity, Third-Party Models do not include third-party provided AI or machine learning models that the AI Services integrate with but that you do not directly use in the AI Services (for example, a third-party AI service that powers an assistant feature).

“Third-Party Offering” means a model, dataset, application, product, service, solution, AI agent, or any other offering by a third party that Benchling makes available for your use directly with the AI Services and that is subject to terms and conditions separate from your Services Agreement, such as an open source license, third party terms, or other terms. For clarity, Third-Party Offerings include Customer Third-Party Services and may include Third-Party Models, and do not include third-party provided AI or machine learning models that the AI Services integrate with but that you do not directly use in the AI Services (for example, a third-party AI service that powers an assistant feature).

“Trials and Betas” means access to the Application Services (or Application Services features) on a free, trial, beta, preview, or early access basis.

2. Opt-In, Authorization, and Agreement

Certain AI Services may require you to enable or initiate them. By doing so, you expressly opt in to the use of these AI Services, authorize us to process your AI Input and other relevant Customer Data through our AI systems to provide the requested functionality, and agree to these AI Terms. You acknowledge that to generate AI Output or take Actions, the AI Services may analyze your AI Inputs and Customer Data, which may involve sending such Customer Data to Benchling’s AI systems or servers or to third-party AI providers (as described in the Documentation).

3. Acceptable Use and Restrictions for AI Services

You agree to use the AI Services in accordance with these AI Terms, your Services Agreement, and the Acceptable Use Policy. In addition to any restrictions on use of the Benchling Services in the foregoing, you agree that you (or any Authorized Users) will use the AI Services responsibly and in compliance with the following rules:

Lawful Purposes Only: You will not use the AI Services in any manner that is unlawful, harmful, or violates any applicable laws or regulations. The AI Services should only be used for legitimate, legal business or personal purposes (within the bounds of your Services Agreement), and may not be relied on for any purpose of obtaining advice or instructions to carry out illegal or dangerous activities.

No Violation of Rights: You will not use the AI Services to infringe upon the rights of others. This includes not providing AI Input that violates intellectual property rights, privacy rights, or other rights of any person or entity, or that is intended, or would reasonably be expected, to generate AI Output that does so. You will not use the AI Services to generate or disseminate content for which you do not have the rights or that contains unauthorized personal data of others without permission.If you input any data or content you obtained from a third party (e.g., uploading a document or including text that you did not create) into an AI Service, you must have the necessary rights to use that content. You are responsible for any third-party content you provide as AI Input and for ensuring that using it with the AI Services (and using any AI Output derived from it) does not violate any law or agreements (such as confidentiality obligations or copyright licenses).

Representations about AI Output. You will not represent any AI Output as being approved or vetted by Benchling.

AI Output Use and Review: You agree to carefully review AI outputs before using them, especially before sharing them with others or making decisions based on them. If the AI Services are used to take Actions on your behalf, you must vet and approve those Actions. You remain solely responsible for any consequences of using AI Outputs.

Competitive Use, Reverse Engineering, and Data Extraction. You will not use any AI Services to develop a similar or competing product or service. You will not reverse engineer or extract any components of an AI Service or its models and algorithms (such as using prompts to discover training data or model weights). You will not attempt to extract, harvest, or otherwise collect data from the AI Services by automated or programmatic means (including web scraping or similar methods), unless expressly authorized in writing; this restriction does not apply to access through our published APIs, SDKs, or other officially supported tools, provided such use complies with Benchling’s Documentation and policies.

Respect System Integrity: You will not attempt to probe, manipulate, or misuse the AI Services beyond their intended functionality. For example, you will not attempt to trick or prompt the AI Services into violating these AI Terms (e.g., no “jailbreaking” the AI with malicious prompts). You will not deploy automated scripts or bots to access the AI Services in a manner that burdens our system or circumvents usage limits, or otherwise engage in any activity that could damage, disable, overburden, or impair the AI Services or the Benchling Services. You will not introduce any viruses, worms, malware, or harmful code into the AI Services or the Benchling Services.

Benchling reserves the right to monitor usage of the AI Services for compliance with the above rules and may suspend, limit, or terminate your access to the AI Services if we believe you are in violation of them.

4. Customer Data Usage and Feedback

4.1 Use of AI Input and AI Output for Service Evaluation, Improvement, and Security

You hereby agree that Benchling may access and use AI Input and AI Output (both of which are Customer Data) to operate, evaluate, improve, and support the AI Services, comply with applicable law, enforce our terms and policies, and keep the Benchling Services safe and secure. However, Benchling will not use any AI Input or AI Output, or any other Customer Data, to train any AI or machine learning models, or permit any third-party service provider to do so, without your permission or instruction, except where you use an AI Service that requires the use of Customer Data to train models specifically tailored for your use, provided that no such models specifically tailored for your use will be used for any other customer or user.

4.2 Feedback

If you provide Feedback to us (through the AI Services or otherwise) regarding any AI Input, AI Output, or Action, we may use that Feedback for evaluation and improvement of the AI Services. An example of such Feedback may be a rating – such as the use of the thumbs up/down icon – for an AI Output or Action in response to an AI Input. You have no obligation to give us Feedback, but if you do, you agree that we may use the Feedback however we choose without any obligation or other payment to you.

5. Intellectual Property

5.1 Ownership of AI Input and AI Output

You retain ownership of your AI Input, and to the extent permitted by law, you will own the AI Output generated in response to your AI Inputs. We hereby assign to you any rights we may have in the AI Output (to the extent they exist under applicable law). You hereby grant Benchling a limited, non-exclusive, transferable, sub-licensable, royalty-free, and worldwide license to host, use, distribute, modify, run, copy, publicly perform or display, translate, and create derivative works of your AI Input and AI Output as per Section 4.1 (Customer Data Usage and Feedback) and as otherwise permitted in these AI Terms and your Services Agreement.

Notwithstanding the foregoing, you acknowledge that AI Output may include or be derived from underlying content or patterns from AI or machine learning training data, which could be publicly available information or third-party content. We do not guarantee that every AI Output is entirely unique, and you acknowledge that AI Output provided to you may be similar or identical to AI Output independently provided to others.

5.2 Benchling’s Rights

All rights, title, and interest in the AI Services remain with Benchling and its licensors. These AI Terms do not grant you any rights to Benchling’s intellectual property or the intellectual property of third-party AI providers, except for your limited license to use the AI Services as provided.

5.3 No Implied Rights

Except as expressly stated in these AI Terms, these AI Terms do not grant either party any rights to the other’s content or intellectual property, by implication or otherwise.

5.4 Third-Party Claims

Due to the nature of the AI Services, Benchling does not represent or warrant that: (a) any AI Output does not incorporate or reflect third-party content or materials; or (b) any AI Output will not infringe third-party intellectual property rights. Notwithstanding any indemnification provisions in your Services Agreement, Benchling does not agree to indemnify you for third-party intellectual property rights claims arising from your use of AI Outputs.

6. Disclaimers and Limitations of Liability

6.1 No Warranties

The AI Services are provided “as is” and “as available” without any warranties of any kind. To the maximum extent permitted by law, Benchling disclaims all warranties, express or implied, regarding the AI Services, AI Outputs, and Actions, including but not limited to warranties of accuracy, completeness, reliability, merchantability, fitness for a particular purpose, or non-infringement, as well as any implied warranty arising from statute, course of dealing or performance, or trade use. Benchling does not warrant that the AI Services will meet your requirements or that the operation of the AI Services will be uninterrupted or error-free.

6.2 Reliance on AI Outputs and Actions

AI Output and Actions are generated by artificial intelligence or machine learning, which are still improving in accuracy, reliability, and safety. When you use the AI Services, you acknowledge and agree:

AI Outputs may not always be accurate and may contain material inaccuracies even if they appear accurate because of their level of detail or specificity.

Actions may not be error free or operate as you intended.

You should not rely on any AI Outputs or Actions without independently confirming their accuracy or correctness.

Benchling makes no warranty or guarantee as to the accuracy, completeness, or reliability of AI Output, or as to the correctness of Actions, and will in no way be liable or responsible for your use of AI Output or any omissions or errors in AI Output or Actions. You are responsible for evaluating AI Output for accuracy and suitability for your use case, and for independently checking the correctness of Actions, including by employing human review of AI Output and Actions.

6.3 Customer Models, Third-Party Models, and Third-Party Offerings

Benchling disclaims all liability arising from your use of Customer Models, Third-Party Models, or Third-Party Offerings with the AI Services, and Benchling’s indemnification obligations do not apply to allegations arising from Customer Models, Third-Party Models, and Third-Party Offerings.

Any Third-Party Offerings (including applicable Third-Party Models) used by you with the AI Services will be governed by your agreement with the third party, not these AI Terms or your Services Agreement. Benchling may, but is under no obligation to, make such third-party agreements available for your review. Benchling does not warrant, support, or assume any liability or other obligation with respect to such Third-Party Offerings or how such third parties use Customer Data.

6.4 Limitation of Liability

In any event, Benchling’s total liability arising from or related to the AI Services (whether in contract, tort, or otherwise) is limited under the same limitation of liability clause as in your Services Agreement.

7. Indemnification

You agree, at your own cost, to defend Benchling from and against any Covered AI Claims (defined below) and will indemnify and hold harmless Benchling from and against any damages or costs awarded against Benchling (including reasonable attorneys’ fees) or that you agree to in settlement resulting from Covered AI Claims. This indemnity is in addition to any indemnification obligations in your Services Agreement.

“Covered AI Claims” are any third-party claims arising out of or related to your use of the AI Services, your AI Inputs, AI Outputs or Actions, or your breach of these AI Terms, including any claims that the AI Output or Actions you directed the AI Services to generate or take violate a third party’s rights or caused harm. This indemnity is in addition to any indemnification obligations in your Services Agreement.

8. Updates to AI Terms

We reserve the right to modify these AI Terms at any time. We will notify you of any material changes by posting the updated AI Terms on our website or through other reasonable means. Your continued use of the AI Services after the effective date of the revised AI Terms constitutes your acceptance of the changes.