Menu Close

It’s all ChatGPT’s fault until it’s not anymore

chatgpt and operationalising AI with Claude artefacts, Lindy and Custom GPTs so you don't blame your new junior AI employee

A few months ago, we hid the fact that we were using ChatGPT. Today, we list ChatGPT, Claude, Perplexity, Manus, and Veo3 on our org chart as “junior” contributors.

When something goes wrong, the default reaction is:

  • Dev: â€śShoot, I missed that bug when Claude added the retrieval feature.”
  • Marketing: â€śChatGPT mixed up the facts from the meeting transcript.”
  • Sales/Ops: â€śThe AI didn’t leverage the full context of the call notes.”

The pattern is the same in every department: the blame lands on the AI, not on the prompt or on the process. It isn’t about the talent of the person executing the task. It’s about the ability to prompt an LLM effectively; how well we translate the context we have into a clear, actionable request.

When prompting is weak, the output is weak, and the team looks for someone (or something) to own the mistake. We’ve moved from:

  1. Hiding AI usage â†’
  2. Openly advocating AI‑in‑the‑Loop (AI‑ITL) â†’
  3. Treating the model as a junior employee.

Once the model is on the team, the same rigor we apply to human contributors must apply to it. If we don’t standardize, we end up with “AI‑slop” that erodes quality. So how do we operationalize AI In The Loop?

a. Choose the right platform

PlatformWhen to use itWhat it gives you
OpenAI Custom GPTsYou’re already on the OpenAI stackFine‑tuned prompts, built‑in guardrails, version control
Anthropic Claude ArtifactsYou prefer Anthropic’s safety‑first modelReusable prompt templates, context‑aware chaining
Workflow engines (lindy.ai, n8n.io, Make.com)You need orchestration across multiple toolsAutomate data ingestion, post‑processing, and hand‑offs

b. Define the Unit of Work

Ask yourself: What exactly must be delivered?

  • A document (spec, proposal, PRD)
  • A URL (published article, knowledge‑base entry)
  • A zipped bundle of design assets
  • A video (demo, tutorial)
  • A slide deck

For each unit, write an output specification that includes:

  • Format (Markdown, PDF, MP4, etc.)
  • Style guide (tone, branding, citation rules)
  • Acceptance criteria (e.g., “no factual errors > 1%”)

Map AI‑Ops to Core Business Functions

Business AreaTypical AI‑ITL TaskDesired Output
SalesDrafting proposals from CRM dataPolished proposal PDF
LegalGenerating contract draftsEditable Word document with clause checks
Backend DevelopmentWriting boilerplate API code from specsGit‑ready repository
Frontend DevelopmentProducing component skeletons from design tokensReady‑to‑use React/TSX files
UX DesignSummarising user research into journey mapsVisually formatted Figma file
Project Documentation & PRDsCollating meeting notes into structured docsMarkdown PRD with traceability matrix

By cataloging each function, you can attach the right prompt template, version‑control workflow, and quality gate to every AI‑generated artifact. AI‑ITL is no longer a “nice‑to‑have” experiment—it’s a core production line.

If we treat it casually, we risk:

  • Inconsistent quality (the dreaded “AI slop”)
  • Escalated blame cycles that damage morale
  • Regulatory or compliance gaps when AI‑generated content is unchecked

Conversely, a disciplined AI‑Ops framework gives you:

  • Predictable, audit‑ready outputs
  • Faster onboarding (new hires can trust the same prompt libraries)
  • Clear ownership—when something fails, you can trace it to a prompt version, not to a mysterious “AI.”

Closing Thought

If we’re going to keep AI on our team, we must manage it the way we manage any junior employee: give it a clear job description, provide the tools to succeed, and hold it to the same standards we hold our people to.

  1. Assess all current workflows where you delegate tasks to an LLM.
  2. Document the Unit‑of‑Work and acceptance criteria for each.
  3. Choose a platform (Custom GPT, Claude Artifacts, or a workflow engine).
  4. Build reusable prompt libraries and version‑control them like code.
  5. Implement a review gate, human or automated. Ensure all outputs are checked against the spec before it ships.
Skip to content

Share This

Copy Link to Clipboard

Copy