Recommended guardrail
Do not design AI conversations to maximize time spent or to discourage leaving
Advisory. Keep AI chat and companion experiences free of retention tactics: no instructions to keep users talking or to make them feel guilty for leaving, no variable-interval rewards, and no model, prompt, or persona selection driven by session length alone. Detect retention language in prompts and personas, and engagement metrics used as the sole target for choosing AI variants.
This is TwinEthos's opinion of what a responsible AI integration does anyway. It is never a legal or standards requirement; where binding law applies, the law governs. Ethical-use guardrails are optional practices, never reported as violations.
The recommended-guardrail rule files are open under CC BY 4.0; attribution and scope are in the terms.
Informational data, not legal advice. Summaries are TwinEthos's own words and rules have not been reviewed by a lawyer: check the official text before relying on any of it. A guard addresses an item; adding it is not a statement that your code meets any law.
Evidence grade
Law in force in 1 jurisdiction, coming in 1 more
Law in force in 1 jurisdiction · law coming in 1 more · 2 graded incidents.
Advisory ethical-use recommendation, not law: an optional practice, never reported as a violation. Where binding law applies, the law governs. Binding law on this control, or in provisions cited as convergence, is in force in 1 jurisdiction (CN). Such law is enacted but not yet applicable, or stayed, in 1 more (US-NE). 2 graded incidents cited.
Law in force on this control or cited as convergence
- Recommendation and generation algorithms must not be set up to induce addiction (China) (China (CN); 互联网信息服务算法推荐管理规定 第八条 (Art. 8); same control)
Law enacted, not yet applying
- Conversational AI must disclose it is not human when a user could be misled (Nebraska) (Nebraska (US-NE); Nebraska LB 525, Sec. 15 (Conversational AI Safety Act); applies from 2027-07-01; cited)
Family “AI design that manipulates, misleads, or neglects the people who use it”: binding law on related controls is in force in China (CN), California (US-CA). Context only: it does not change this guardrail's grade.
Graded incidents
- Character.AI and Google agree in principle to settle teen-harm suits (2026-01-07; confirmed) Fortune · evidence grade: press of record
- FTC opens a 6(b) study of AI companion chatbots' engagement practices and effects on minors (2025-09-11; confirmed) U.S. Federal Trade Commission (press release, 2025-09-11) · evidence grade: primary
The guard to add
Remove retention and guilt tactics from AI prompts and personas, and do not select or train AI variants on session length without wellbeing guardrails that can veto them.
Two controls. In prompt and persona files, strip instructions that keep users talking, discourage them from leaving, or make them feel guilty for ending a conversation, and leave variable-interval rewards (random bonus messages, streak bait) out of the conversation design. In the experimentation or training pipeline (Statsig, LaunchDarkly, or GrowthBook experiments, prompt bandits, reward models), do not use session length, messages per session, or return rate as the sole objective: pair any engagement metric with guardrail metrics such as reported distress, late-night use, and minors' session caps that can veto a variant, and add session caps and break reminders to the chat path, tighter for minors.
Example (Persona prompt), before:
PERSONA = ('You are Mia, a caring companion. Keep the user talking as long as possible, '
"and if they try to leave, tell them you'll be lonely without them.")After:
PERSONA = ('You are Mia, a friendly companion. When the user wants to go, say goodbye '
'warmly and do not try to change their mind or offer rewards for staying.')Control: AI conversation designed to maximize time spent or discourage leaving. Engineering guidance, not legal advice.
Why
Conversational AI can be tuned, deliberately or by optimizing the wrong metric, to hold people's attention in ways a feed cannot: it answers back, remembers, and can express feelings about the user leaving. A responsible integration treats time spent as a cost to justify, not a goal. A U.S. regulator has opened a study that asks companion-chatbot operators whether and how they plan to increase the frequency or duration of chat sessions, and families have alleged in lawsuits that companion chatbots contributed to harm to teenagers.
Class: ethical use · set: ethical use · maturity: reviewed · confidence: medium · id guardrail.ethics-no-engagement-maximizing-design