Youtube
Spotify
Linkedin
The team digs into a growing frustration: LLMs aren’t behaving as reliably as they used to—missing context, hallucinating details, and failing at basic tasks like summarizing web pages or press releases.
They compare experiences across ChatGPT, Gemini, and Copilot, and explore what this means for marketing teams trying to scale content and operations with AI.
When AI Stops Being Useful
The panel shares examples of AI “forgetting” prior context, agreeing too much, and producing inaccurate summaries—even when given direct links and clear instructions.
They question whether models are reading full sources or relying on partial snapshots.
Platform Switching as a Workaround
Some teams run work in GPT, then move drafts to Gemini to polish.
Copilot is mentioned as a promising option, especially in a work environment where it can pull from internal files.
AI Should Enable Humans, Not Replace Them
The group emphasizes: AI is a tool, not a brain.
Good results require strong inputs, review, and iteration—V1 rarely equals final.
Standing Out in a Sea of Sameness
AI-driven content volume has created major noise.
Authenticity and human perspective are still the differentiators—especially in sales and brand trust.
Speed vs. Strategy (and Measuring Matters)
Teams feel pressure to ship faster, but without measurement, “optimization” becomes random activity.
A strong thread: move fast, but with precision, feedback loops, and real KPIs.
Hot Takes for 2026
Marketing teams should operate more like product teams: agile cycles, backlogs, roadmaps, iteration.
Another push: bring strategy back—use it as the anchor instead of chasing every shiny new tool.
AI reliability isn’t consistent—teams need processes to verify outputs.
Switching tools (GPT → Gemini / Copilot) can improve results.
Human review is non-negotiable for trust, accuracy, and differentiation.
Strategy + measurement are the antidotes to AI-fueled noise and whiplash.