Closed
Close: Lethal Trifecta does not qualify as a semantic anchor#123
Conversation
Co-authored-by: rdmueller <1856308+rdmueller@users.noreply.github.com>
Copilot
AI
changed the title
[WIP] Add Lethal Trifecta anchor proposal
feat: Add Lethal Trifecta semantic anchor
Feb 24, 2026
The previous implementation described 'Lethal Trifecta' as a microservice coupling anti-pattern (temporal/behavioral/data coupling) — which is NOT what Joy Heron means by the term. The actual concept relates to dangers of AI assistants (see: https://www.innoq.com/de/blog/2026/02/dangers-of-ai-assistants/). Ironically, an AI assistant hallucinated a confident, plausible-sounding but completely fabricated definition — demonstrating the very phenomenon the actual 'Lethal Trifecta' likely describes. The anchor is removed until the correct definition can be sourced from the original blog post." Co-authored-by: rdmueller <1856308+rdmueller@users.noreply.github.com>
Copilot
AI
changed the title
feat: Add Lethal Trifecta semantic anchor
revert: Remove hallucinated Lethal Trifecta anchor
Feb 24, 2026
Co-authored-by: rdmueller <1856308+rdmueller@users.noreply.github.com>
Copilot
AI
changed the title
revert: Remove hallucinated Lethal Trifecta anchor
feat: Add Lethal Trifecta semantic anchor (Joy Heron, INNOQ)
Feb 24, 2026
The term fails the Consistent quality criterion: an LLM hallucinated a completely wrong definition (microservice coupling) when asked, proving the term is not yet established in LLM training data. Joy Heron coined it in a February 2026 blog post — too recent to be reliably activatable. The discussion itself served as a live demonstration of why this term does not meet the bar for inclusion in the catalog. Co-authored-by: rdmueller <1856308+rdmueller@users.noreply.github.com>
Copilot
AI
changed the title
feat: Add Lethal Trifecta semantic anchor (Joy Heron, INNOQ)
Close: Lethal Trifecta does not qualify as a semantic anchor
Feb 24, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
"Lethal Trifecta" (Joy Heron, INNOQ, Feb 2026) fails the Consistent quality criterion: the term is too new to exist in LLM training data, so models hallucinate plausible-sounding but wrong definitions. This PR's own history demonstrates the problem — the first implementation confidently described a microservice coupling anti-pattern that has nothing to do with the actual concept.
A term that causes LLMs to confabulate is the opposite of a semantic anchor.
Changes
docs/anchors/lethal-trifecta.adocandlethal-trifecta.de.adocdocs/all-anchors.adocdocs/metadata/roles.ymlwebsite/public/data/*.json(52 anchors)Original prompt
💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.