Acceptable Use & Platform Rules
These Rules apply to prompts, messages, shared documents, generated content, hosted projects, integrations, automated actions and go-to-market activities. Violations may result in content removal, feature restrictions, project suspension or account termination. They apply together with our Terms of Service, which incorporate them by reference.
Prohibited uses
The following are expressly prohibited by the Company, and any user deemed by the Company to be engaging in such activities is subject to content removal, feature restrictions, project suspension, and/or account termination:
- ·Illegal activity, facilitation of crime, sanctions/export evasion, trafficking, exploitation or instructions intended to cause unlawful harm.
- ·Malware, credential theft, phishing, unauthorized access, destructive code, botnets, denial-of-service or security-control circumvention, except authorized defensive testing within permitted scope.
- ·Harassment, threats, stalking, doxxing, non-consensual intimate content, sexual exploitation, child sexual abuse material or material facilitation of abuse.
- ·Infringement or misappropriation of copyright, trademark, trade secret, privacy, publicity, confidentiality or other rights.
- ·Fraud, impersonation, deceptive practices, fake reviews, fabricated evidence, misleading endorsements, or AI-generated material presented as authentic where materially misleading/harmful.
- ·High-impact decisions in employment, housing, education, credit, insurance, legal services, health or essential services without lawful authority, meaningful human review and safeguards.
- ·Unlawful discrimination or targeting based on protected characteristics, including advertising/automated decision systems.
- ·Bulk unsolicited messaging, spam, unlawful robocalls/texts, unlawful lead generation, scraped-contact outreach or circumvention of consent/anti-abuse controls.
- ·Use creating material physical-safety, critical-infrastructure, weapons or severe-harm risks except legitimate, authorized protective/research/compliance uses.
- ·Reverse engineering model weights, extracting confidential system data, bypassing safety/rate controls, or using Services to train a competing foundation model except as expressly permitted.
- ·Using the Services to violate connected third-party terms, including ad platforms, app stores, cloud services, social networks or model-provider rules.
UGC and community safety
Where users can post, share, message or collaborate, Company may provide reporting, blocking, access-control and moderation tools. You may not abuse reporting systems. Company may remove content or restrict users who violate these Rules.
Automated and agentic actions
Use agentic features only within accounts, files, systems and data you are authorized to access. You are responsible for permissions, limits, supervision and backups. Do not instruct an agent to send communications, publish content, spend money, enter contracts, delete data or modify production systems without authorization and appropriate approval controls.
Enforcement and appeals
We may investigate suspected violations using automated systems and human review, preserve evidence and take proportionate action. Where appropriate and legally required, we will provide notice and an appeal method. We may act without advance notice for urgent safety, security, legal or abuse risks.
Reporting and appeals in the product
Every public content surface carries a Report action; reports are reviewed promptly. If you believe we removed your content or restricted your account in error, contact [email protected] with a link to the content and a short explanation, and we will take another look.