For more than a century, the court reporter has occupied a peculiar throne in the machinery of justice: the person who captures every word, every stammer, every objection, translating the chaos of human speech into an official record at speeds exceeding 225 words per minute. It is a profession built on an arcane skill—stenography—that takes years to master and demands a particular kind of concentration that most humans simply cannot sustain. Now that throne is being measured for a different occupant.
Artificial intelligence, specifically the class of speech-to-text systems that have improved dramatically in recent years, is infiltrating courtrooms, deposition suites, and hearing rooms across the legal world. The shift is not sudden or dramatic. It is happening in the way most technological displacement happens: gradually, then all at once.
The economics of listening
The math is straightforward and unforgiving. Training a certified court reporter takes two to four years of specialized education, with pass rates for certification exams historically hovering below fifty percent. The profession has faced a persistent shortage for decades, with courts in some jurisdictions canceling proceedings simply because no reporter was available. Meanwhile, AI transcription services can be deployed instantly, scaled infinitely, and operated at a fraction of the cost per hour.
This does not mean the technology is flawless. Courtrooms present particular challenges: overlapping speakers, heavy accents, technical legal terminology, poor acoustics, and the occasional witness who mumbles into their chest. Human reporters have always navigated these obstacles through experience, intuition, and the simple act of asking someone to repeat themselves. AI systems are improving rapidly on these fronts, but they still stumble where humans adapt.
What gets lost in translation
The court reporter's value has never been purely mechanical. Experienced reporters develop an understanding of legal procedure, an ear for what matters, and a professional judgment about when to interrupt for clarification. They serve as a kind of quality control on the record itself. When a witness contradicts earlier testimony, a skilled reporter notices. When an attorney's question is genuinely unintelligible, a reporter flags it.
AI transcription systems, by contrast, produce text without comprehension. They render sound into words with increasing accuracy but without any sense of meaning, context, or legal consequence. The question facing the legal profession is whether that distinction matters enough to justify the premium for human reporters, or whether good-enough transcription at lower cost will win the day.
The hybrid future
The most likely near-term outcome is neither full automation nor the preservation of the status quo, but a hybrid model already emerging in some jurisdictions. AI handles the initial transcription; human editors—often former or current court reporters—review and correct the output. This preserves some human oversight while capturing the efficiency gains of automation. It also transforms the nature of the work from real-time performance to post-hoc quality control, a fundamentally different skill set with different compensation structures.
For working court reporters, this transition presents a familiar dilemma: adapt to a supporting role alongside the machines, or exit the profession entirely as the economics shift beneath them.
Our take
The stenographer's decline is not a tragedy of technology run amok. It is the latest chapter in a very old story about specialized human skills becoming economically unviable in the face of good-enough automation. What makes court reporting worth watching is that it sits at the intersection of technology and justice, where the stakes of getting the record wrong are not merely commercial but constitutional. The legal system will eventually decide how much it values human judgment in the transcription process. That decision will say something important about how we weigh efficiency against the intangible qualities that humans bring to consequential work.
