This AI Disclosure Statement (“Statement”) explains how artificial intelligence is used in the Surely application and related services (the “Services”), the limitations of that technology, and the safeguards Surely Networks, LLC (“Surely Networks,” “Surely,” “we,” “us”) applies. This Statement is incorporated into, and forms part of, the Surely Terms of Use.
1. Where AI Is Used
- The Surely assistant and harness. A conversational assistant and underlying processing system (the “harness”) that help you capture, organize, search, understand, and act on the information in your brain.
- Skills, routines, and agents. Configurable automations that can summarize, draft, extract, classify, schedule, and take other actions on your behalf, including on a recurring or scheduled basis.
- Ingestion and organization. Automated systems that read, summarize, classify, link, and organize the data you connect — such as email, calendar entries, documents, and meeting transcripts — to build and maintain your brain.
Content generated by these systems is referred to in the Terms of Use and this Statement as “AI Output.”
2. AI Output Is Not Human-Reviewed by Default
AI Output is generated automatically by machine-learning systems, including large language models operated by Surely’s technology providers. Unless the Services expressly state that specific content has been reviewed and approved by a human being, you should assume that AI Output has not been reviewed by any person before it is shown to you or acted upon.
3. Limitations and Accuracy
AI Output may be incomplete, outdated, or factually incorrect, and may contain plausible-sounding but false statements sometimes called “hallucinations.” AI Output does not constitute legal, financial, tax, medical, or other professional advice, and is not an official statement of any person or organization unless a human has reviewed, approved, and adopted it as their own.
You are solely responsible for independently verifying any AI Output before relying on it, acting on it, or forwarding it to a third party.
4. How Your Data Is Processed by AI
Surely processes the data you provide and connect in order to operate the Services and produce AI Output for you. As described more fully in our Privacy Policy:
- We do not sell your personal information;
- We do not use the content of your brain or your connected data to train foundation models; and
- Our AI providers and other subprocessors are contractually restricted to using your data only to provide services to Surely, and are prohibited from using it to train their own general-purpose models where we can obtain that contractual protection.
How this works today, unless your license says otherwise. The description above reflects how Surely processes data today, when Surely operates the harness and the cloud infrastructure on your behalf. If you use Surely under an Enterprise License Agreement, that agreement (and any Data Processing Addendum) governs and may provide for different arrangements — for example, storing data in your own environment or using your own models (“bring your own harness”). In case of conflict about data handling for enterprise customers, the Enterprise License Agreement controls.
5. Safety and Human-Review Protocols
These protocols describe our general practices; they do not create a guarantee that any particular AI Output has been reviewed or is error-free.
- Content guardrails. Our AI systems are configured with guardrails intended to block or limit outputs that promote violence, harassment, unlawful discrimination, self-harm, or illegal activity.
- Human-in-the-loop for higher-risk actions. Certain features are designed so that a person makes the final decision. Where an automation, routine, or agent would take a sensitive or irreversible action, Surely is designed, where offered, to seek your confirmation before acting.
- Pre-deployment testing. Before material changes to our AI features are released, we review prompts and representative outputs — including adversarial or “red-team” testing — for accuracy and safety.
- Ongoing monitoring and escalation. We monitor system performance and user reports on an ongoing basis. Content flagged as potentially unsafe or seriously inaccurate may be routed for review and may result in prompt adjustments, additional filtering, or feature changes.
- Incident response. If we identify AI Output that causes, or is reasonably likely to cause, material harm, we work to correct or remove it promptly and to address the underlying cause.
- Third-party AI providers. Our AI features rely on large language models operated by third-party providers, currently OpenAI, L.L.C. and Anthropic, PBC. When you use AI features, your prompts and the content reasonably necessary to fulfill your request are made available to these providers to process it, and that processing is subject to those providers’ applicable terms and conditions and to our agreements with them. Those agreements are intended to restrict how your data may be used, including prohibiting the providers’ use of your content to train their own general-purpose models where we can obtain that contractual protection, as further described in the Privacy Policy.
6. Automated Actions and Agents
Surely can be configured to take actions on your behalf — for example, through routines, scheduled tasks, agents, or connected integrations. You are responsible for the automations you enable and their scope, including any messages sent, files changed, or actions taken by an automation you configure. Review the permissions and scope of any automation before you enable it.
7. Your Responsibilities
- Treat AI Output as a starting point for your own judgment and work, not as a final authority;
- Do not represent AI Output as professionally reviewed, authoritative, or human-authored unless it has, in fact, been reviewed and adopted by a qualified person; and
- Do not attempt to manipulate, “jailbreak,” or misuse Surely’s AI systems to generate content that violates the Acceptable Use & Community Guidelines.
8. Reporting Concerns and Feedback
- How to report. Email ai@surelynetworks.com or use an in-product reporting control where available.
- What to include. Please provide as much detail as possible, including the content in question, where and when it appeared, and why you believe it is inaccurate, unsafe, or otherwise concerning.
- Acknowledgment. We aim to acknowledge reports submitted through these channels within 30 business days.
- Use of feedback and privacy. Feedback you provide may be used to improve the Services. Please do not include sensitive personal information in a report. Information you submit is handled in accordance with the Privacy Policy.
9. Relationship to Other Documents
This Statement supplements, and should be read together with, the Terms of Use and the Privacy Policy. In the event of any conflict specifically regarding disclaimers of AI accuracy, this Statement controls; in all other respects, the Terms of Use controls (and, for enterprise customers, the Enterprise License Agreement controls as to matters it addresses).
10. Changes to This Statement
We may update this Statement from time to time as our AI systems and safeguards evolve, consistent with the Terms of Use. The most current version will always be available through the Services.
11. Contact
Questions about this Statement can be sent to ai@surelynetworks.com or to Surely Networks, LLC, 2810 N Church St, PMB 548750, Wilmington, Delaware 19802-4447.