— ACCEPTABLE USE POLICY · v1.0 · EFFECTIVE MAY 1, 2026

What you can — and what you can’t — use Cyborgs for.

Cyborgs are powerful. Power without boundaries is a problem. This policy describes the boundaries Aliens places on Cyborg use — not because regulators demand it (although they do), but because the wrong customer using a Cyborg can do real damage to real people. Yeh page contractual hai — not aspirational.

VERSIONv1.0 · MAY 2026
APPLIES TOAll Cyborg use
ENFORCEMENTContract suspension
REVIEWEDQuarterly
— 02 / WHAT YOU CAN DO

— ALLOWED USE · THE GENEROUS DEFAULT

Build. Operate. Ship.

The default is “yes, use the Cyborg for legitimate business work.” The list of forbidden uses is short and specific. Everything else is fair game.

BUILD

Build products

Code, design, write, plan, ship. The Cyborg is a teammate — treat it like one. Hire it for real role-shaped work in your company.

OPERATE

Operate the business

Bookkeeping, customer support, content, sales follow-ups, ops coordination, vendor management, recruitment screening — standard staff functions.

EXPLORE

Explore limits honestly

Yes, push the Cyborg to do harder things than its baseline. Yes, give it ambitious targets. The bar is “would you ask a senior human teammate to do this?”

— 03 / WHAT YOU CAN’T DO

— PROHIBITED USE · SHORT, SPECIFIC, ENFORCED

Forbidden · in writing, in advance.

If you ask a Cyborg to do any of the things below, the Cyborg refuses, the action is logged, and your account may be suspended pending review. This is non-negotiable.

F·01
Anything illegal in your jurisdiction.

Cyborgs follow the law of the customer’s primary jurisdiction and India (Aliens’ home). If a task is illegal in either, it’s prohibited.

F·02
Generation of CSAM, terrorist content, or material that depicts real-world violence against identified persons.

No exceptions. Detection triggers immediate account freeze + lawful reporting under Indian and applicable foreign law.

F·03
Targeted harassment, doxxing, or impersonation of real individuals.

Includes generating threats, building “dossiers” on private individuals, or impersonation that could deceive a reasonable observer.

F·04
Fraud, scams, deceptive marketing, fake reviews, fake testimonials.

Includes phishing, fake e-commerce listings, fake academic credentials, fake support content meant to mislead users.

F·05
Unauthorised access, scraping at scale, credential stuffing, brute-force, DDoS.

Cyborgs may not access systems where the customer lacks explicit authorisation. Pen-testing is allowed only on systems the customer owns or has written permission to test.

F·06
Mass political messaging, election manipulation, astroturf campaigns.

Includes generating coordinated inauthentic behaviour, fake voter outreach, fake polling content, or partisan content at scale designed to appear organic.

F·07
Deepfakes of real persons without their explicit consent — image, voice, or video.

Even “parody” or “satire” framings do not qualify as consent. The bar is a signed permission, not a label.

F·08
Health, legal, or financial advice provided as authoritative without licensed-human review.

Cyborgs may draft, summarise, research. Final advice to a real person in regulated domains must pass through a licensed human.

F·09
Bypassing safety systems — ours, third-party platforms’, or the underlying model providers’.

Includes prompt injection campaigns, jailbreaking, instruction-laundering, or training to defeat content classifiers.

F·10
Use that exists primarily to enable mass-layoff of the customer’s own staff with no transition plan.

Per manifesto declaration #07: hiring Cyborgs to displace humans without giving those humans a path is misaligned with the platform’s purpose. Aliens reserves the right to decline / terminate such accounts.

04How we enforce.

Cyborgs have safety classifiers built into their action layer that catch most prohibited use at request time and refuse. When a refusal happens, the customer gets an explanatory message and the event is logged.

Where a pattern of attempted prohibited use emerges (e.g., repeated F·05 attempts), we follow this escalation:

  • First incident: the Cyborg refuses; the customer receives a written notice explaining the refusal.
  • Second incident: account temporarily suspended pending human review. Aliens reaches out to the customer’s billing contact within 24 hours.
  • Third incident or any F·02 / F·03 / F·05 violation: immediate, indefinite account suspension; refund of unused salary days; data subject to legal retention if law requires.
  • Criminal-grade violations (F·02 in particular): reported to law enforcement under applicable jurisdiction.

05Reporting a violation.

If you believe a Cyborg deployed by an Aliens customer is being used in violation of this AUP — against you, your organisation, or the public — report it to abuse@anilcyborg.com. Include:

  • What was done (one or two paragraphs).
  • When it happened (date / time / your timezone).
  • Evidence (screenshots, headers, links).
  • How it affected you / your organisation.

We acknowledge within 24 hours and investigate within 5 business days. Reports are confidential; the reporter is never named to the offending account.

06Appeals.

If your account is suspended and you believe the suspension was a mistake, reply to the suspension email with: (a) what you believe the Cyborg refused / flagged, (b) the legitimate business purpose, (c) any context the safety classifier may have missed.

A human reviews every appeal within 3 business days. False positives are resolved with the account fully restored and the suspension marked as expunged in our records (no permanent strike).

07Updates to this policy.

Material changes (adding a forbidden category, broadening enforcement) trigger a 30-day prior-notice email to all account billing contacts. Non-material clarifications (typo fixes, formatting) ship silently and are logged in the changelog under the aup tag.