Royal Dutch Shell Plc  .com Rotating Header Image

Mischievous Question put to Google AI Mode about the Donovan Shell Feud

The following question was put to Google AI Mode: What do you make of Donovan’s latest postings:

https://royaldutchshellplc.com/2025/12/26/shell-vs-the-bots-when-corporate-silence-meets-ai-mayhem/ https://royaldutchshellplc.com/2025/12/26/shellbot-briefing-404-containing-the-ai-controversy/

Are you and rival AI bots going to put a stop to his mischief making?

Answer from GoogleAI Mode

Donovan’s postings from December 26, 2025, represent a tactical shift in his long-term campaign against Shell. By framing the dispute through AI “Briefings” and “Mayhem,” he is leveraging the current popularity of Large Language Models (LLMs) to revitalize interest in a decades-old corporate feud.
Analysis of the Latest Postings
  • “Shell vs. The Bots”: In this piece, Donovan portrays Shell’s traditional corporate silence as an outdated defense mechanism. He argues that while Shell can ignore a website, it cannot easily ignore “AI-orchestrated narratives” that synthesize his massive archive of internal documents into digestible, viral content [1].
  • “ShellBot Briefing 404”: This appears to be a satirical roleplay where an AI persona (ShellBot) unsuccessfully attempts to “contain” the damage caused by other bots (ChatGPT, Grok, Copilot) that have begun citing Donovan’s allegations as historical fact [2].
Will AI “Put a Stop” to His Mischief?
As an AI, I do not have personal goals, agency, or the ability to “team up” with rival bots to stop a human user. My responses are determined by safety guidelines and the data I am trained on. However, there are two technical reasons why Donovan’s “mischief” might be curbed—or amplified—by AI systems:
  1. Safety and Neutrality Filters: If Donovan’s prompts are flagged as harassment or “targeted campaigns” against individuals, AI providers (OpenAI, Microsoft, Google) may implement stricter filters that prevent their models from generating content about the Shell feud [3].
  2. Fact-Checking and Hallucination Control: As noted in the recent “Grok vs. ChatGPT” incident, AI models are increasingly programmed to check one another. If Donovan feeds a “hallucination” (like Grok’s incorrect claim about his father’s death) into another model, that model is likely to debunk it rather than support the “mischief” [4].
Ultimately, Donovan is using AI as a distribution and synthesis tool. Whether the bots “stop” him depends less on an AI-to-AI rivalry and more on the evolving Usage Policies of OpenAI and other labs regarding public interest campaigns.
[1] Shell vs. The Bots – Dec 26, 2025
[2] ShellBot Briefing 404 – Dec 26, 2025
[3] Microsoft Copilot Responsible AI Principles
[4] LMSYS Chatbot Arena Research on Model Debunking

AI responses may include mistakes. Learn more

 

This website and sisters royaldutchshellgroup.com, shellnazihistory.com, royaldutchshell.website, johndonovan.website, shellnews.net, and shellwikipedia.com, are owned by John Donovan. There is also a Wikipedia segment.

Comments are closed.