User Preferences Now Drive AI Output

Version: 1.10.0, Date: September 17, 2025

User preferences now actually change what the AI produces. Previously, when you set preferences for tone, format, length, or context (like industry, audience, or timeline), those settings were mentioned in the generated prompt but not enforced. The AI would often default to generic corporate formatting regardless. Every preference you set is now translated into specific behavioral instructions that measurably change the AI’s output. The rest of this post is grouped by app area.

Across the App

• Fixed: Preferences are now enforced, not just mentioned. Previously, selecting a tone like “Creative” or a format like “Mixed” resulted in those words being passed through without actionable guidance. The AI would see “Tone: creative” but default to standard corporate prose anyway. Each preference now maps to explicit behavioral instructions (for example, “Write in flowing, engaging prose. Avoid corporate jargon.”) that the AI actually follows.
• Fixed: Format settings are now respected. If you select a specific output format preference, the generated prompt now includes concrete format enforcement instructions instead of a generic label. This eliminates the default bullet-point formatting that used to appear regardless of your selection.
• Improved: Dynamic context enforcement. Context fields like industry, timeline, audience, and key metrics are now processed through a detection system that generates targeted enforcement instructions. If you specify a healthcare industry context, the prompt instructs the AI to use healthcare-appropriate terminology and frameworks, not just mention that the industry is healthcare.
• Improved: Scalable preference handling. The enforcement system uses pattern-based detection rather than fixed mappings, meaning it handles any input you provide rather than only recognizing a fixed list of options. Custom or niche preferences are translated into useful instructions just as effectively as common ones.