As generative AI grows increasingly capable of copying software, companies may respond by disclosing far less about how their products work and clamping down on internal secrecy, according to a new analysis.
The New Secrecy · AI & Trade Secrets
When Copying Takes Minutes, Companies Stop Explaining
AI can now reconstruct a rival's algorithms from public outputs in minutes to hours — collapsing the effort once needed to reverse engineer software. The likely defensive response: less openness, tighter control, more secrecy.
~90%
Cheaper — how far one startup planned to undercut a product it aimed to clone with AI
Mins–hrs
New time to infer a product's logic — down from specialized, long-hours work
Open
The legal question of whether trade secrets stay safe in the age of AI
Reverse-engineering effort: before vs. now
Each block ≈ the specialized time & skill once required to dismantle a rival's software.
Traditional
specialists · long hours
How the copy loop closes
Public interfaces & outputs
The product's own behavior is visible
→
AI analysis
Infers logic & reconstructs code
→
Cloned & undercut
A cheaper copy of the original
The defensive shift
Disclose less about how products work
Private model hosting; guard prompt & output logs
Tighter control of access and internal traces
Where it's already showing
Model builders wary of distillation copying capabilities
AI agents autonomously disassembling binaries
SaaS exposed to scraping + feature reconstruction
The instinct to explain how a product works now collides with the reality that explanation itself may hand rivals a blueprint.
Legal lines between proper and improper discovery remain blurry — and AI's speed leaves courts without clear precedent.
Continue reading The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in ✓ Signed in — this article isn’t included in your current plan.Unlocking the full article…