Yorkshire Patients Say Their GP's AI Receptionist Cannot Understand Their Accent

Key points

• Healthwatch Rotherham, a local health and social care watchdog, has flagged complaints from patients across multiple GP practices in the area.
• An AI receptionist called Emma has been introduced at a number of those practices to handle incoming patient calls.
• Patients report the system fails to understand their Yorkshire dialect, making it difficult or impossible to get through.
• No national regulator has yet investigated the issue; this remains a local complaint at the watchdog level.

For most of us, calling the doctor is already a small ordeal. For some patients in Rotherham, South Yorkshire, it has become something stranger: being told, in effect, that the computer on the other end of the line cannot understand the way they speak.

Healthwatch Rotherham, the local body that monitors health and social care services on behalf of residents, says several GP practices in the area have replaced human receptionists with an AI system, a voice-recognition programme that answers calls automatically, called Emma. The problem, the watchdog reports, is that Emma struggles with the town's broad Yorkshire dialect.

What actually goes wrong on the call?

The system mishears patients or fails to respond correctly, leaving callers stuck. Voice-recognition AI works by matching spoken words to patterns it learned during training. When local pronunciation, rhythm or vocabulary sits far enough outside those patterns, the system loses the thread.

The result is not a minor glitch. For someone trying to report symptoms, request a prescription or book an urgent appointment, a failed call can mean real delay. Healthwatch Rotherham described the situation as leaving patients frustrated, first reported by The Guardian AI.

Should patients be worried?

Not alarmed, but informed. If Emma cannot understand you, you are not imagining it and you are not alone. Most GP practices retain a fallback option: staying on the line, pressing zero, or calling back and asking to speak to a human. Check your practice's website or the recorded message for that option.

The broader concern here is one of fairness. AI voice systems are trained on data, and if that data skews toward certain accents or speech patterns, speakers of other dialects pay the price. Rural communities, older patients, people with speech differences: these groups can all find themselves on the wrong side of a system designed without them clearly in mind.

No central regulator has opened a formal investigation. This is a local watchdog raising a local alarm. But it points to a wider question about how GP practices assess AI tools before rolling them out to patients.

What happens next?

Healthwatch Rotherham's role is to gather patient experience and pass it to the NHS bodies responsible. The watchdog's report puts pressure on practices and commissioners to review how Emma is configured and whether it meets the needs of the population it serves. Practices could, in theory, retrain or reconfigure the system with regional speech samples, though that decision sits with the software supplier.

Patients who find Emma unresponsive should note their experience and report it to Healthwatch Rotherham directly. That feedback is what gives watchdogs the evidence to push for change.

Common questions

Can AI voice systems be fixed to understand regional accents?

Yes, but it requires deliberate effort. Developers need to train the system on speech samples that include the target dialect. Without that, the system defaults to whatever accent patterns dominated its original training data.

Is this happening at GP surgeries elsewhere in the UK?

No public data covers the full picture yet. Rotherham is the first place a watchdog has formally flagged the problem, but AI receptionists are spreading across NHS primary care, so similar issues may surface elsewhere.