Generative artificial intelligence can enable Indians — even those who do not have a technical background — to solve the country’s most pressing problems, according to Mercedes Soria, executive vice-president and chief intelligence officer at US robotics security company Knightscope. “Now, all we have to worry about is, are we solving the right problem,” said Soria, who was on an India visit to share her expertise on machine learning and AI.
Soria is a speaker for the United States Department of State on technology and women in STEM (science, technology, engineering, mathematics). “In the future, it’s not going to be so much that you have to be technical to have a startup and solve a problem. It’s going to be more of, do you identify the right problem to solve, and then the technology like Gen AI will help you,” she said.
Problems can be solved even by those who can’t write a single line of code, she added. This is a ‘massive opportunity’ for India with its large population and unique local issues. Young people in India, including high schoolers, are very familiar with and using tools like ChatGPT, whereas in the US this is seen more among those in university, Soria noted.
While countries can cooperate to advance AI skills and address the talent shortage, the technology can help tackle existing problems, as nocode platforms are emerging to enable this, Soria said. At the same time, learning is being driven by large companies themselves. “I know that a lot of Indian people aim to go to the US and study there and work there. But what I’m trying to communicate to everyone is you don’t have to do that — the technology is available to you where you are,” Soria said.
Future thinking
On US President Joe Biden’s AI executive order, which among other things seeks to attract global AI talent, Soria said any country, not just the US, now aspires to have the best talent. Yet, the perceived impact of the US elections on spending decisions of businesses may be overstated, according to Soria.
“I’m not going to say there’s absolutely no impact… There will be some regulations and some law that could benefit a company or another, but in the grand scheme of things, I don’t think it is going to completely change the landscape of what’s been happening with large companies right now,” Soria said.
“Yes, there’s a dependency of companies on what they can and cannot do depending on who wins the election, but I don’t think it’s such a destabilising power as a lot of people make it out to be,” she added.
Further, Soria said that big companies like Microsoft and Apple have already made their bets in terms of how they’re going to invest in Gen AI, and what is emerging now is how much time it will take for them to pay off the capital expenditure that they invested.
Waiting game
What is proving to be a challenge, however, is the disruptions in the global supply chains, and in particular the scarcity of graphics processing units (GPU). Notably, the US has put in place certain export controls on critical AI-related components amid tensions with China, its rival in the AI race. Soria’s company Knightscope combines self-driving technology, robotics and AI to predict and combat crime. It recently ordered about 3,000 GPUs and faced a six-month waiting period. “I don’t want to speak for the US government, but my personal opinion is that some of those restrictions need to be lifted,” Soria said.
“The world is not going to be able to withstand that sort of waiting, so there’s going to have to be some give and take somewhere, where some of those components are allowed to come into the US or other countries.” Countries’ ability to meet their sustainability commitments will also become more difficult, Soria said, given the amount of energy required in Gen AI ‘server farms’.
“This wasn’t something that was planned — three years ago we didn’t know we were going to have hundreds of thousands of servers in just one computer room,” she said. “But I trust that humans will find a solution around it, if not powered by the same software that we are creating, it could help us find a solution.”
- 0 Comments
- Ai Process
- Artificial Intelligence