Union Minister Accuses Gemini-AI of Spreading False Information About PM Modi

Central Minister of Electronics and Technology Rajeev Chandrasekhar has cautioned Google India about the reactions generated by AI regarding Prime Minister Narendra Modi. Rajeev Chandrasekhar stated that Google has violated the IT Act’s regulations and the Criminal Code.

Rajeev Chandrasekhar has reshared a post about X, highlighting how Google’s AI provides misleading answers to questions about certain global leaders, including PM Modi.

“Google has violated many provisions of the IT Act and many provisions of the Criminal Code,” wrote Rajeev Chandrasekhar in his post. He further stated, “‘Direct violation of section 3(1)(B) of these IT Act rules and many provisions of the Criminal Code are also violated.'”

Google’s AI is facing challenges with the verification of historical facts. Instead of rectifying many parts of history being misrepresented, Google’s AI is perpetuating inaccuracies. Many users of Google’s AI have been sharing historically inaccurate content on social media platforms.

Google has stated that they are working to address these issues faced by their AI. They mentioned that they have been working on resolving these problems within their image generation feature. In the meantime, they will halt image generation and release an improved version shortly.

Google launched its new Artificial Intelligence (AI) model, Gemini, three months ago to compete with ChatGPT. This AI tool operates much like humans. Google claims that Gemini performs tasks such as understanding, reasoning, coding, and planning better than other models.

The company’s CEO, Sundar Pichai, stated that this marks the beginning of a new era of AI at Google. Gemini is Google’s latest Large Language Model (LLM). Pichai provided information about it for the first time at the developer conference in June.

Gemini is based on the Massive Multitask Language Understanding (MMLU) model. The model’s ultra variants are adept at reasoning and image understanding, performing better in 30 out of 32 benchmark tests compared to ChatGPT 4. Gemini Pro outperforms GPT 3.5, the free version of ChatGPT, in 6 out of 8 benchmark tests.

Large Language Models (LLMs) receive training from large datasets. These models, also known as deep learning algorithms, utilize large datasets for training. They are capable of translation, inference, and generating text and other content. LLMs are recognized as neural networks (NNs) and are inspired by human brain-driven computing systems. LLMs can understand protein structures, write software code, and perform many other tasks through training.