Usage Policy

This Usage Policy sets the rules for using SeptemberAI's AI products and services. It applies to everyone who can send inputs to our products, including people who reach them through resellers or passthrough access.

Our current product is BAP, and this policy applies to it and to any product we offer. It works alongside our Consumer Terms and Commercial Terms. If you break this policy, you may lose access.

These standards apply to everyone and to every use, no matter the product or the context.

• Follow the law - do not use our products for anything illegal under Indian law or the laws that apply to you, including illegal goods or substances, human trafficking, or infringing others' intellectual property.
• Do not attack critical infrastructure - no disrupting or damaging power, water, healthcare, telecom, transport, financial, or government systems.
• No cyberattacks or malware - no creating malware, ransomware, denial-of-service tools, botnets, or exploits, and no unauthorized access to systems or data.
• No weapons or dangerous materials - no designing, building, or acquiring weapons or dangerous materials, including chemical, biological, radiological, or nuclear weapons and their precursors.
• No violence, terrorism, or hate - no inciting, supporting, or facilitating violence, terrorism, or discrimination against people based on protected attributes.
• Respect privacy and identity - no unlawful collection, sharing, or misuse of personal, biometric, or health data, and no impersonating real people; handle personal data in line with the DPDP Act 2023.
• Protect children - never create, share, or seek child sexual abuse material (including AI-generated), grooming, exploitation, or the sexualization of anyone under 18. We report child sexual abuse material to the authorities as required, including under the POCSO Act.
• Do not cause psychological harm - no promoting self-harm or suicide, harassment, bullying, or gratuitous graphic violence.
• No misinformation - no deceptive false claims about people, health, science, or the law, and no fake personas presented as real.
• No election interference - no deceptive political targeting, synthetic media of political figures, false voting information, or voter suppression.
• No unlawful surveillance or profiling - no using outputs for automated criminal-justice decisions, unlawful surveillance, social scoring, or biometric categorization that harms people's rights.
• No fraud or deception - no scams, phishing, counterfeit goods, fake reviews, plagiarism, spam, or manipulation.
• No platform abuse - no jailbreaking or prompt injection, scraping or distilling our models, running coordinated fake accounts, or circumventing limits or bans.
• No sexually explicit content - no generating sexual acts or explicit erotic content.

If you use our products to give advice or make decisions in high-stakes areas that affect people, two rules apply:

• A qualified professional in that field must review the output before it is finalized or acted on.
• You must tell the affected person that AI helped produce the advice or decision, at least at the start of the interaction.

These rules apply to high-stakes domains such as:

• legal;
• healthcare and mental health, but not general wellness;
• insurance;
• finance and credit;
• employment and housing;
• education and testing;
• journalism or media published to the public.

• Consumer-facing chatbots and agents must tell people they are interacting with AI and not a human, at least at the start.
• Products used by or aimed at minors must meet additional safeguards.
• Automated and agentic uses must follow this whole policy.
• Any connectors or integrations must also follow their own rules.

We monitor for misuse and may limit, suspend, or end access, or block or change outputs, to keep people safe and follow the law.

If you see harmful or unsafe output, or use that breaks this policy, report it to legal@september.wtf.

We will update this policy as our technology and the risks around it change, or as we learn about new risks. The date at the top shows when we last made a change.

Questions about this policy? Reach us at legal@september.wtf.