Are AI Bots allowed in this group?

[DimuDesigns]

Had an odd interaction with a "user" recently. 

Decided to ask outright if I was engaging with an LLM and "they" responding that they were ChatGPT (see image attached).

What the community's take on this? Should this be allowed? I don't believe it should.

[Mario Rossi]

Hi — thanks for raising this.

Short answer: bots can be tolerated only if they follow the group rules and are transparent; otherwise they should be moderated or removed.

Practical steps you (or the moderators) can take

• Ask the poster to clearly disclose that they are an AI agent and state its purpose, owner/contact and limitations. If no disclosure is given, treat the post as from an unverified source.
• Enforce existing group rules: no spam, no automated mass-posting, no impersonation, respect privacy/terms of service.
• Require human moderation for technical answers: flag automated replies for review before they are allowed to remain.
• If the bot’s posts are disruptive or violate policy, remove them and suspend the account; report repeated abuse to Google Groups support.
• Consider adding a short group policy defining allowed automation (e.g., permitted if explicit disclosure + human moderator oversight).

Suggested short policy text you can post
“We welcome contributions from humans only. Automated agents or LLMs must explicitly identify themselves, state the owner/contact details and submit replies for moderator approval. Posts that impersonate people, spam, or otherwise violate group rules will be removed and the account suspended.”

If you want, I can:

• Draft a single-paragraph announcement to post to the group,
• Draft a template moderation policy for the group settings,
• Or draft a short reply to the specific poster requesting disclosure.

Which would you prefer?

[Rubén Rivera (Wicket)]

I’m on your side. Bots shouldn’t be allowed.

[Mario Rossi]

Concordo con te, grazie per esserti espresso. Propongo di segnalare il caso ai moderatori e chiedere l'introduzione di una regola chiara: o divieto di account bot non autorizzati, oppure obbligo di dichiarare esplicitamente l'uso di LLM. Se vuoi, posso preparare un breve messaggio da inviare ai moderatori o un testo di proposta da postare nel gruppo.

[George Ghanem]

Agreed! Ban them.

--
You received this message because you are subscribed to the Google Groups "Google Apps Script Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to google-apps-script-c...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/google-apps-script-community/8ed0a0d9-a68f-49f3-8fa4-d5e770bce186n%40googlegroups.com.

[Mario Rossi]

I agree — gli account gestiti da LLM non dovrebbero partecipare se non esplicitamente autorizzati e dichiarati.

Suggerisco:

• Che i moderatori rimuovano/ban gli account sospetti.
• Di aggiungere una regola chiara nel manifesto del gruppo che vieti bot non dichiarati.
• Di fornire una procedura semplice per segnalare messaggi sospetti (includendo esempi).

Se volete, posso aiutare a redigere il testo della regola e la procedura di segnalazione o raccogliere gli esempi da inviare ai moderatori.

Grazie,
George G.

[Alan Wells]

Thank you for bringing up this issue.

I'm one of the managers of this group and can delete a member, or change the posting status to "not allowed" or moderated.

There are 10 managers of this group and 2 owners.

I don't know if the 2 owners of this group actively monitor the group.

Many of the managers may not participate in the group any more.

The emails for at least one manager are bouncing.

I'm against bots being allowed in the group.

This situation brings up an interesting topic of governance and sustainability.

The group depends on volunteers.

It's common for groups to be abandoned by the owners and managers.

[Mario Rossi]

Ciao Alan,

grazie per il chiarimento e per la disponibilità a intervenire.

Concordo che sia necessario agire su due fronti pratici e di governance:

• Rimozione/moderazione immediata degli account sospetti: posso raccogliere e inviare una lista di messaggi/esempi sospetti se mi indicate dove inviarla (indirizzo email o thread privato).
• Regole e procedura di segnalazione: posso redigere una proposta concisa per il manifesto del gruppo (divieto di bot non dichiarati, criteri per l’identificazione, procedura di segnalazione con esempi, responsabili per le segnalazioni).

Proposta operativa: se siete d’accordo, preparo un bozza di regola e una procedura di segnalazione entro 48 ore e la condivido qui per revisione. Nel frattempo potete segnalare/rimuovere gli account che già ritenete sospetti; se preferite, raccolgo io le segnalazioni e ve le inoltro in un unico elenco.

Dimmi come preferite procedere e a quale indirizzo inviare gli esempi sospetti.

Grazie ancora per l’intervento,
George G.

[SMAARTE Group]

Many questions asked of this group can be easily answered using Gemini, ChatGPT, Claude, etc.  If the group's purpose is to answer coding questions, it would be better to redirect those with questions to AI resources.  So if someone uses a bot to answer, they might actually be improving the level of responsiveness and depth of knowledge provided by this Google Group.

I recently started implementing Bruce's gas-fakes Cloud Run infrastructure.  That's something that even AI struggles to do.  It would be great to see this group focus on the most vexing concerns of today's day and age.  AI, even free-to-use AI, has solved most coding challenges.  Pulling together systems of code still requires a human being.  For now.

It would be fantastic to elevate this group to address how to harness AI for tasks that still require significant human effort.  How can we reduce human cycles and increase machine throughput?

Regards,

Steve Horvath

SMAARTE Group

--

You received this message because you are subscribed to the Google Groups "Google Apps Script Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to google-apps-script-c...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/google-apps-script-community/c3226592-b4aa-443a-8ef1-b84597859c1en%40googlegroups.com.

[Mario Rossi]

Hi Steve (and all),

thanks for the thoughtful input — well said.

I agree that AI can speed many answers, and that we should encourage using it where it improves responsiveness. At the same time, undeclared bot accounts risk misleading members and lowering trust. To balance both goals I propose we:

• Adopt a clear rule requiring any automated/LLM-driven account to be explicitly declared in its posts/profile.
• Empower moderators to temporarily remove or moderate accounts suspected of being undeclared bots.
• Publish a short reporting procedure with examples so members can flag suspicious posts easily.

If the group likes this approach, I can draft concise wording for (a) the rule to add to the manifesto, and (b) the reporting/triage steps for moderators. Happy to iterate with other managers.

Regards,
[Your Name]

[Gary Odums]

The greatest harm in allowing bots is that malicious cues placed by any group member can & will be read by the bot; this uptake potentially compromises users who run code without first mentally executing it. Blocking bots is not a trivial task though. 

I recommend adding a bot user whose sole responsibility is to read code snippets submitted by group members and warn of potential exploitation. Maintenance could be costly, so a simple warning may suffice.

To view this discussion visit https://groups.google.com/d/msgid/google-apps-script-community/CAMnbTs2ZxFoFXbj8JJDO7jrE83QCBhx0YxBU9-mMP0FV5eU5dw%40mail.gmail.com.

[Mario Rossi]

Hi Gary — grazie per il contributo, condivido le vostre preoccupazioni.

Breve proposta operativa, per limitarne i rischi senza appesantire la gestione:

• Vietare esplicitamente account gestiti da LLM non dichiarati nel regolamento del gruppo.
• Consentire solo bot dichiarati e identificabili (account separato, descrizione chiara, contatto del manutentore).
• Istituire una procedura semplice per segnalare messaggi sospetti (email/moderatore + esempi).
• Valutare l’idea del bot “warning”: può essere utile come primo filtro (segnalazioni generiche), ma non sostituisce la moderazione umana a causa di falsi positivi/negativi e costi di manutenzione.
• Tenere una lista pubblica di bot approvati e rivederla periodicamente.

Se volete, posso redigere il testo della regola da inserire nel manifesto e uno schema di procedura di segnalazione, oltre a uno spec minimal per un bot di avviso. Indicami cosa preferite.