JAKARTA – The Indian government has warned technology companies developing new artificial intelligence (AI) tools that their products must be approved by the government before being made public.

According to an alert issued by the Indian IT Ministry on March 1, this approval must be given before the public release of “unreliable” AI tools or those still in the testing phase, and these tools must be labeled for possibility of providing inaccurate answers. to the questions.

“Availability to users of the Indian Internet must be ensured with the explicit permission of the Government of India,” the Indian IT Ministry said.

Additionally, the alert asks platforms to ensure their tools do not “threaten the integrity of the electoral process” as elections are expected to take place this summer.

The new warning comes shortly after one of India's top ministers criticized Google and its AI tool, Gemini, for “inaccurate” or biased responses, including one that Indian Prime Minister Narendra Modi had been described by some as “fascist”.

Google apologized for Gemini's shortcomings and said the tool “may not always be reliable,” especially for today's social topics.

“Security and trust are legal obligations of platforms. 'Sorry, unreliable' does not trump the law,” Rajeev Chandrasekhar, India's deputy IT minister, said in a tweet on Platform X.

In November, the Indian government announced it would introduce new regulations that would help contain the spread. deep fakes generated by AI before the next election – a measure also implemented by regulators in the United States.

However, Indian officials have received objections from the tech community over its latest AI warning, saying India is a leader in technology and that it would be a “crime” if India ” deviate from this leadership.”

Chandrasekhar responded to this “noise and confusion” in a follow-up article on X, saying there should be “legal consequences” for platforms that “allow or outright produce illegal content.”

“India believes in AI and is fully committed not only to talent but also to the expansion of our digital and innovation ecosystem. India’s ambitions on AI and ensuring safe and reliable internet for internet users are not binary,” Chandrasekhar said.

He also explained that the warning is only intended to “guide those who use untested or under-tested AI platforms to the public Internet” to be aware of the obligations and consequences under the Indian law and the best way to protect yourself and users.

Tag: artificial intelligence India artificial intelligence