Something shifted in early 2026. A Harvard Business Review study gave a name to what strategists had been sensing but could not quite articulate. When you feed a strategy problem to an LLM — any LLM — it does not analyze your organization. It predicts the most socially desirable response, drawn from the same pool of contemporary management thinking that every other organization is drawing from. The researchers called it Trendslop.
The study tested seven large language models across thousands of simulations. Every model, regardless of the company context provided, gravitated toward the same answers: differentiate over commoditize, augment over automate, explore over exploit, long term over short term. When pressed to choose differently, the models resisted. When allowed to choose both, they did — producing what one researcher described as trying to be everything at once, dressed in a suit and pretending to be sophisticated.
Better prompting shifted the bias by roughly twenty-two percent. Richer context shifted it by eleven. Neither fixed it.
The reason is not the tool. It is the process the tool is being applied to.
Most organizations cannot clearly articulate how their long-term strategy is actually created — beyond a five-box diagram and a familiar sequence of retreats. That absence of process clarity is not a minor gap. It is the reason AI keeps returning the same answers. There is nothing sufficiently specific, sufficiently rigorous, and sufficiently honest about the organization’s own situation for the model to work against. So it defaults to the internet average. Every time.
LTSP26 exists because this is a solvable problem — and because the solution is not more AI literacy. It is process clarity first, AI application second.
The conference runs September 15 to 17, 2026, entirely virtually, across three half-days designed for active participation rather than passive viewing. The content draws directly on practitioners who have applied process-driven AI interventions inside real strategy engagements — not theorists, not futurists, not vendors with platforms to sell.
The argument at the center of everything: before any organization can benefit from AI in its strategy work, it needs an honest, detailed picture of how that work currently happens. Not a deck. Not a process diagram. A rigorous, stage-by-stage understanding of what is actually occurring — where the thinking is sharp, where it is foggy, and which stages are genuinely vulnerable to Trendslop and which are not.
That baseline is what makes AI useful. Without it, applying AI to strategy does not improve the process. It accelerates whatever the process already is — fog and all.
LTSP26 is where that argument gets examined, tested, and put to work.