DeepSeek, a Chinese AI startup, has faced criticism following a NewsGuard audit that revealed its chatbot achieved only 17% accuracy in delivering news and information. The audit ranked DeepSeek's chatbot tenth out of eleven competitors, trailing behind Western rivals such as OpenAI's ChatGPT and Google's Gemini. The chatbot exhibited a fail rate of 83%, repeating false claims 30% of the time and providing vague or unhelpful answers in 53% of cases. This performance has raised concerns about the reliability of DeepSeek's technology in comparison to its Western counterparts.
DeepSeek's chatbot achieves only 17% accuracy rating in NewsGuard audit, trailing Western competitors https://t.co/EmZM0kn2q0
🤖 El chatbot de la startup china de IA #DeepSeek logró solo un 17% de precisión en la entrega de noticias e información en una auditoría de NewsGuard.https://t.co/Y7eDNe6oCj
DeepSeek's chatbot achieves 17% accuracy, trails Western rivals in NewsGuard audit https://t.co/J9n7cr56gq