AI Video Interview Bias: What It Scores in Neurodivergent People @NikiKnobel
AI Video Interview Bias: What It Scores in Neurodivergent People  @NikiKnobel
Uploaded July 2026 | Updated November 2026, 3 days ago
AI Video Interview Bias: What It Scores in Neurodivergent People

The AI video interview is one of the most common tools in modern algorithmic hiring, and almost nobody applying is told how it actually works. This short explains it. After legal pressure on facial analysis, most of these systems stopped scoring your face and started scoring a text transcript of your answer. Your audio is recorded, transcribed into text, and the natural language of that transcript is scored against competencies. So your result depends entirely on whether the machine captured your words correctly, and on whether you produced enough of them.

For autistic and ADHD job seekers, that design carries specific AI hiring bias risk. Unusual pacing, dysfluency, a flat or atypical prosody, and speech under heavy processing load all make transcription less accurate. An AI video interview cannot score words it did not capture. There is a second problem that hits many neurodivergent people even harder. An external audit found that some candidates gave answers that did not contain enough content for the algorithm to generate a meaningful score at all. The system issued no warning and did not prompt anyone to say more. It simply returned a low result.

Think about what that means for an autistic person who answers the exact question that was asked, precisely, and then stops. In an ordinary human conversation that is efficient and clear. To an AI video interview scoring transcript length and content, a short answer does not read as efficient. It reads as almost nothing. This is where AI recruitment bias and neurodivergence collide, because the very communication style that makes many autistic and ADHD people excellent at their actual work is the style the measuring instrument struggles with most.

Researchers also tested whether rewriting the interview questions to be clearer and more accessible closes the gap. It helps, and on its own it does not close it, because the gap is produced by what the model rewards rather than by what the question asks (Benson et al., 2025). That is why accommodations still matter in algorithmic hiring, and why knowing how the AI video interview works changes what you can ask for.

In this short I give you the one practical habit that protects you most in a recorded interview, plus the questions worth asking an employer before you ever hit record. If you want the legal grounding across five jurisdictions in plain language, the free toolkit is linked for you as well.
AI Video Interview Bias: What It Scores in Neurodivergent PeopleADHD at Work: Rethinking Task Switching & DeadlinesADHD Employment Tribunal Cases  Discrimination ExplainedBody Doubling: Youre Already Doing It!Master Your Goals  Create Your If Then Plan Today!ADHD in Women: Dismissed and Misdiagnosed by DoctorsUnlock Your Brains Reward System: Dopamine & ExplorationAutistic and Sick: Why I Never Notice Until I Crash | Ramblings of an Autistic ADHDerAutistic Sleep: Your Experience Matters, Trackers DontThe Workplace Doom Pile: Where Tasks Go to Get LostNovelty Bonus: Unleash Your Brains Hidden Power! #4AudHD Sleep: Understanding Your Body Clock & Nervous System
Nicola Knobel (Audhd) |

AI Video Interview Bias: What It Scores in Neurodivergent People

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER