Biometric Authentication Still Needs an Accessible Fallback, Researcher Argues

Biometric authentication has become a familiar way to unlock devices and reach sensitive accounts, but a newly published analysis argues the technology is only as inclusive as the fallback it offers when a face or fingerprint scan fails. Writing for Biometric Update on October 7, UX and accessibility researcher Manisha Varma Kamarushi said that for blind and low-vision users, the fallback experience often determines whether they can reach a device, financial account, workplace system, or healthcare portal independently.

The article notes that biometrics can reduce friction by removing the need to type long passwords, visually locate controls, or complete multiple interaction steps. But fingerprints may fail to register, and facial authentication may fail because of lighting, positioning, or environmental conditions; some users also choose not to use biometrics at all. When that happens, the user is pushed into a fallback method.

In many digital products, that fallback is still a conventional PIN or password interface. Kamarushi writes that for blind and low-vision users a traditional touchscreen keypad depends heavily on spatial awareness and accurate interaction with on-screen controls, and that the experience can still be slower, more cognitively demanding, or less private even when a screen reader makes the interface technically accessible.

That problem led to her work on OneButtonPIN, an authentication method designed to let blind and low-vision users enter a PIN using a single button rather than a visual keypad. She argues that accessibility is strongest when it is built into the interaction model itself rather than added after the core design has already been established.

The analysis challenges the idea that accessibility necessarily weakens security, reframing the goal as resilience: a system that offers multiple secure and usable authentication pathways is better equipped to handle failure and reduces dependency on a single method. An accessible fallback, she writes, should not automatically be slower, harder, or less secure because a user cannot use the primary method.

Kamarushi calls on security and product teams to ask accessibility questions earlier in the design process, including whether an authentication flow can be completed non-visually, whether users can recover independently from a failed biometric attempt, and whether the fallback works well with assistive technology. As the industry moves toward a passwordless future, she concludes, that future also needs to be accessible by design, because the fallback experience can be the difference between independent access and exclusion.