Global AI Speech Recognition Models Fail Indian Languages, Impacting Critical Services

The article explicitly involves AI systems—automatic speech recognition models from global providers like OpenAI, Microsoft, Google, and Meta. The use of these AI systems in voice interfaces for critical services means their malfunction (high word error rates and poor recognition of dialects) directly leads to harm such as miscommunication in welfare applications, healthcare, and other essential services. This constitutes indirect harm to persons and communities through disruption or degradation of critical infrastructure services. The harm is realized, not merely potential, as the article details the performance failures and their practical consequences. Hence, the event meets the criteria for an AI Incident rather than a hazard or complementary information.

The article explicitly involves AI systems (speech recognition models from OpenAI, Meta, Microsoft, and Sarvam). It discusses their use and performance shortcomings in Indian languages, which could plausibly lead to harm in critical domains like welfare and medical applications due to transcription errors. However, no actual harm or incident is reported, only the potential for harm due to high error rates. Hence, it fits the definition of an AI Hazard rather than an AI Incident or Complementary Information. It is not unrelated as it clearly concerns AI systems and their impact.

The article explicitly involves AI systems—automatic speech recognition models—and discusses their development and use in real-world Indian contexts. It identifies significant performance deficiencies that could plausibly lead to harm, such as miscommunication in critical services accessed via voice interfaces. Although no direct harm has been reported yet, the described performance gaps and their implications for digital inclusion and service access constitute a credible risk of harm. The article's focus is on revealing these risks and encouraging improvements, not on reporting an actual incident or harm caused by AI. Hence, the event fits the definition of an AI Hazard, as it plausibly could lead to AI incidents if the issues remain unaddressed.