Enjamb Labs policy evolution
Before/after stance changes across captured policy versions, with exact citations. If no before/after delta is available yet, AIRIN shows the latest citation-backed stance events instead.
No before/after stance delta is available for this filter yet. Latest citation-backed stance events are shown below.
Aug 28, 2026traininghigh
data sharing
Latest stance: sale or sell
“This Privacy Policy describes how Enjamb Labs, Inc., a Delaware corporation (“Enjamb,” “we,” “us,” or “our”), handles personal data in connection with enjamb.ai, app.enjamb.ai, and the software, APIs, agents, and related services we provide (together, the “Services”). Enjamb is enterprise software for biopharma research and development. Organizations enter into a written agreement with us, we provision a workspace for them, and their administrators invite named users. There is no consumer sign-up and no self-serve plan. Three commitments sit behind everything below. We do not sell personal data. We do not train models on customer content. An organization's data belongs to that organization, and we hold it to deliver the Services and for no other purpose.”
Open citationAug 28, 2026traininglow
model training
Latest stance: no training claim
“This Privacy Policy describes how Enjamb Labs, Inc., a Delaware corporation (“Enjamb,” “we,” “us,” or “our”), handles personal data in connection with enjamb.ai, app.enjamb.ai, and the software, APIs, agents, and related services we provide (together, the “Services”). Enjamb is enterprise software for biopharma research and development. Organizations enter into a written agreement with us, we provision a workspace for them, and their administrators invite named users. There is no consumer sign-up and no self-serve plan. Three commitments sit behind everything below. We do not sell personal data. We do not train models on customer content. An organization's data belongs to that organization, and we hold it to deliver the Services and for no other purpose.”
Open citationAug 28, 2026traininglow
model training
Latest stance: no training claim
“The Services use large language models, some operated by us and some by model providers we engage as subprocessors. When an agent runs, the content it needs is sent to the selected model to produce the result, and is retained only as long as the run and its audit record require. We do not train, fine-tune, or otherwise improve any model on customer content or on the personal data described in this Policy, and our agreements with model providers prohibit them from doing so with data we send on a customer's behalf. Where a provider's terms differ for a particular model, that model is presented as an explicit choice with its terms stated, and it is not enabled by default. Agents act with the authority of the person who asked. An agent reaches a connected system as that user, inside the permissions that user already holds, so it cannot open a study, a document, or a dataset the requester could not open themselves. Actions that change a system of record can be placed behind human approval by an administrator. Model output can be wrong. The Services are built so that a result carries the sources, actions, and assumptions behind it, and they are intended to support the judgment of a qualified professional rather than replace it. Nothing the Services produce is medical, legal, or regulatory advice.”
Open citationJul 20, 2026traininglow
model training
Latest stance: no training claim
“The Services include AI-assisted research tools. Enjamb does not operate a proprietary foundation model; we route requests to third-party model providers (see Section 5.2 for the current list). 4.1 No training on your private content. We do not use Your Content (including AI Inputs and Outputs) to train Enjamb-owned AI models. Enjamb does not own foundation models. Our third-party model providers contractually commit, under the enterprise API tiers Enjamb uses, to not train their models on traffic sent through Enjamb's API access. If a provider's policy materially changes, we will update this Policy and notify active users. 4.2 Inference only. Inputs are sent to model providers only to generate the response to your specific request. We retain transcripts of your AI conversations within your Project so you can revisit them; you can delete those transcripts at any time. Model providers may retain API logs for a limited period to enforce their own usage policies and detect abuse, per their published policies. 4.3 You are responsible for AI Outputs. AI Outputs are probabilistic and may be incorrect or unsuitable. Verify them before relying on them, especially for medical, legal, financial, or safety-critical decisions.”
Open citationJul 20, 2026traininglow
model training
Latest stance: no training claim
“4.1 How it works. The Services include AI-assisted research tools. These features rely on large language models and other AI systems operated by third-party providers (see Section 6). Enjamb does not operate its own proprietary foundation models; we orchestrate and route to third-party models on your behalf. 4.2 Inputs and Outputs. "Input" means content you submit to AI Features (prompts, attached files, selections, and related context). "Output" means content returned by AI Features (text, drafts, structured data, figures). Both are part of Your Content under these Terms. 4.3 No training on your private content. Enjamb does not use Your Content (including Inputs and Outputs) to train Enjamb AI models. Our third-party model providers contractually commit, for their enterprise API tiers that Enjamb uses, to not train their models on API traffic from Enjamb. If this position changes for a specific provider, we will update these Terms and notify active users. 4.4 AI limitations and your responsibility. AI Outputs are probabilistic and may be incorrect, incomplete, biased, or otherwise unsuitable. You are solely responsible for reviewing, verifying, and deciding whether to rely on any Output. Do not use Outputs as authoritative medical, legal, financial, or safety advice without independent verification. 4.5 Autonomy and tool use. The Services' AI agent may read, write, and modify files within your Project on your behalf, including editing documents, running search queries against third-party databases and websites, generating figures, and other workspace actions. You acknowledge these actions are initiated by your instructions to the agent and are attributable to your Account.”
Open citationGenerated from live stance events. Informational only, not legal advice.