Strike the right balance for the regulation of AI
The Telecom Regulatory Authority of India’s (Trai) recommendation to set up an independent statutory Artificial Intelligence and Data Authority of India, a body that will act as a regulator and advisor for all AI-related sectors, is a welcome step. First, it acknowledges that AI’s impact is being felt not just in telecom, but across sectors like banking, financial services, healthcare, transportation, education and agriculture, to name a few, which calls for a ‘holistic’ regulatory approach. Second, Trai’s 141-page report underscores the connectedness of technologies like 5G and the internet of things (IoT) with AI, which explains why policies can no longer deal with these in silos.
Third, Trai attributes its delay in offering advice—it took two years to submit its recommendations to the department of telecom—to the pace at which AI is evolving, which poses special challenges of regulation. Ever since OpenAI publicly released its large language model-powered chatbot ChatGPT last November, developments in the generative AI space have both excited and unnerved people. Some AI experts including Elon Musk have called for a six-month moratorium on building such foundation models, while others have equated the risk of AI to that of nuclear war.
Others like Yan LeCun and Andrew Ng insist that AI is far from becoming sentient, and that its benefits far outweigh perceived risks. AI, meanwhile, is already in our smartphones, cameras, driverless cars, low-cost satellites, chatbots and robots. It is even helping farmers by providing flood forecasts and warnings, among other things.
But it has a dark side too. Think of AI-generated clones, deep-fakes, malware and other mischief. AI can plagiarize, pose security and privacy risks,
. Read more on livemint.com
