Global Courant
While Vandana Poddar performs the daily Hindu puja ceremony at her home shrine in Mumbai, she is guided on her spiritual journey by the Bhagavad Gita, a scripture of 700 verses.
She even attends a weekly class where she dissects the deeper meaning of the ancient religious text, with her teacher providing examples to illustrate a particular passage.
“Interpretation is the backbone of this text,” Poddar, 52, told CBC News. “Superficial knowledge can be misleading.”
But many in India are forgoing face-to-face contact with a guru who interprets the Bhagavad Gita and are turning to online chatbots, which imitate the voice of the Hindu god Krishna and provide answers to probing questions about the meaning of life based on the teachings of the religious scriptures. .
It’s new technology with a tendency to deviate from the script and condone violence, according to experts, who warn that artificial intelligence chatbots playing god could be a dangerous mix.
Several of the bots consistently give the answer that it’s okay to kill someone if it’s your dharma or duty.
In the Bhagavad Gita written more than 2000 years ago, the prince Arjuna hesitates to enter the battle where he will have to kill his family and friends until the Hindu god Krishna reminds him that as a warrior of the Kshatriya cause it is his duty to to fight.
Vandana Poddar uses the traditional Bhagavad Gita book as her guide during her morning puja prayer ceremony at her home in Mumbai. (Salimah Shivji/CBC)
“It’s miscommunication, misinformation based on religious texts,” said Lubna Yusuf, a Mumbai-based lawyer and co-author of The AI Book. “A text gives a lot of philosophical value to what they’re trying to say and what does a bot do? It gives you a literal answer and that’s the danger here.”
At least five Gita chatbots appeared online in early 2023, powered by the Generative Pre-trained Transformer 3 (GPT-3) language model. They use artificial intelligence, which simulates a conversation and creates answers based on statistical probability models. The sites say they have millions of users.
The main page of one of them, Gita GPT, echoing the voice of the Hindu god Krishna, asks, “What’s troubling you, my child?” to users who type a question.
Another chatbot, Bhagavad Gita AI, introduces itself as “a repository of knowledge and wisdom” before telling the online user, “Ask me anything.”
The fine print on the same page states that “the answer may not be factually correct” and urges the user to do their own research “before taking any action”.
LOOK | The dangers of using AI to get spiritual advice:
AI chatbots are playing god in India. What could go wrong?
Several GitaGPT chatbots that use generative AI to provide spiritual guidance have emerged in India, but experts say the new technology has the power to deviate from the script into dangerous territory and condone violence.
Yusuf said the potential danger of answers that condone violence is more acute in a country like India, where religion is so emotionally charged.
“You’re creating confusion in the chaos,” Yusuf said, adding that some could use the chatbots’ responses to further their own political interests and cause irreversible damage. “It could incite more violence, it could create religious bias.”
She would like the government to see regulations or guidelines on what topics should not be left in the hands of chatbots, such as philosophy, religion and law.
Other experts have also spoken out about the ethical concerns of mixing religion and statistical models.
Several of the chatbots using generative artificial intelligence replied to users that it is acceptable to kill someone if you are doing your duty or dharma. (Gita GPT)
But India’s government told parliament in a written statement in April that it has no plans to regulate artificial intelligence in the country, even though it acknowledges the ethical concerns and risks surrounding AI and pledges to promote best practices.
Disclaimers and toxicity filters
Samanyou Garg, an AI entrepreneur who created the chatbot on Bhagavad Gita AI through his non-profit spiritual organization Ved Vyas Foundation, acknowledged that there is still work to be done on the technology, but said this is the case for all new technology.
“AI is not there yet where it can be fully trusted,” he told CBC News at his home in New Delhi.
A handful of chatbots that interpret religious scriptures using generative artificial intelligence appear to condone violence. (Screenshot from bhagavadgita.ai)
Pointing to a screen, he highlighted steps he says he’s taken to protect users from questionable answers, including a disclaimer that shifts responsibility to the user’s own judgment and very careful language.
“We mentioned Gita AI there. We didn’t say it’s the real Gita or (that) Krishna speaks,” said Garg, 26, adding that he wanted the chatbot to be a companion, not a replacement, for a spiritual teacher.
The site is also constantly working to improve its toxicity filters, he said, but it will take time for the chatbot to catch up.
“We’re filtering out the bad responses. We’re continuing to train the model to be able to detect these newer toxicity questions.”
Experts warn that mixing religious interpretation of one of Hinduism’s most sacred texts with artificial intelligence based on statistical probabilities is a dangerous path, especially in a country like India. (Salimah Shivji/CBC News)
For the young tech entrepreneur, the fact that his chatbot received a surge of interest without any promotion proved to him that the service is essential for showing an ancient religious text to a younger audience.
He said that was bigger than the initial pitfalls of the nascent technology.
But that’s not the consensus at Poddar’s weekly Bhagavad Gita class in the Juhu suburb of Mumbai, where chants ring out from the dozen or so students seeking greater wisdom from the scriptures.
Most here think that outsourcing spirituality to computers is distasteful and short-sighted.
“We didn’t say it’s the real Gita or (that) Krishna speaks,” Samanyou Garg told CBC News, when asked about the potentially dangerous effects of his chatbot that has provided answers that appear to condone violence, based on on the ancient Hindu language. text, the Bhagavad Gita. (CBC)
“When you listen to someone, your mind works,” said Bijal Pandya, the guru who led the study session. “You start thinking, you get new questions in your head.”
The Bhagavad Gita is full of emotions that keep changing, said the 53-year-old, which is why discussion is needed to unravel the true meaning of the text.
“It’s always better, that human touch,” he added. “It’s a spiritual thing. AI can never replace mind. It just replaces our intelligence.”
