The academic world is currently walking a very thin line between protecting research integrity and maintaining basic fairness. As journals rush to deploy new screening tools to catch automated content, a troubling pattern has emerged in 2026. These forensic algorithms often rely on "predictability" and "linguistic complexity" to flag AI writing. Un