---
id: "question-protecting-proprietary-data"
type: "open-question"
source_timestamps: ["§ Applying Gen AI to Proprietary Data"]
tags: ["data-security", "competitive-intelligence"]
related: ["concept-ai-strategy-inference", "contrarian-proprietary-data-moat"]
resolutionPath: "Research into adversarial AI techniques designed to mask strategic operational data from external pattern-recognition models."
sources: ["spine"]
sourceVaultSlug: "hbr-seg-spine"
originDay: 1
articleStem: "hbr-cl-96-ai-no-sustainable-advantage"
sourceUrl: "https://hbr.org/2024/09/ai-wont-give-you-a-new-sustainable-advantage"
sourceTitle: "AI Won’t Give You a New Sustainable Advantage"
---
# Can proprietary datasets ever be truly secured against AI inference?

**Open question.** Even if a company secures its data against breaches, the authors argue advanced AI can observe market outcomes and *infer* the underlying proprietary data and strategy (see [[concept-ai-strategy-inference]]). It remains unresolved whether any obfuscation technique can protect a successful strategy from being reverse-engineered by an AI analyzing public results.

**Resolution path:** Research into adversarial AI techniques designed to mask strategic operational data from external pattern-recognition models.

**Enrichment:** The inference threat is conceptually grounded but empirically thin — there is limited evidence of LLMs systematically reverse-engineering complex corporate strategies from public data at scale. Connects to the broader debate in [[contrarian-proprietary-data-moat]] about whether proprietary data is a strong or weak moat.