Underscoring need for regulations, US-based expert says AI has potential to be catastrophic

BOSTON, MASSACHUSETTS: Boston-based Shabbi S. Khan, Partner Foley and Lardner, LLP, said that the global community has been slow in the past when it responded to the nuclear bomb threat and it is so far doing the same with generative Artificial Intelligence (AI), which also has the potential to be catastrophic.

“AI cannot be ignored and has great upsides, however, it comes with a fair share of challenges that demands cautiousness,” he said while speaking with international journalists at his office in Boston.

The journalists’ meeting with the intellectual property lawyer was a part of the international reporting tour organised by the US Department of State’s Foreign Press Centre.

“The role of regulations in technology is to ensure the responsible and safe development, deployment and use of technological innovations. But you always run the risk of trying to under regulate versus over regulate, and it has its own sort of impacts.”

Khan said global policy harmonisation – identical or similar regulations in the world especially for the US and the European Union in AI – is very important for tech companies.

“Regulations have played a role in important global technologies. For instance, the nuclear proliferation treaty was enacted in 1970, 25 years after the Hiroshima bombing. Clearly, it took a really long time for the global community to band together and say this is a problem.”

“(But yes) It (nuclear proliferation treaty) certainly stopped the proliferation - only a few countries were able to obtain nukes (after that),” Khan told Business Recorder.

“Regulations are necessary but we have to balance them out in the right way (in order not to hinder technological advancement). Global harmonisation will allow for a level playing field.”

Khan was critical about the US’s relaxed approach to AI regulations.

“The US, quite frankly, is waiting,” he said. “And it’s waiting for a variety of reasons. I think, politically, the environment just makes it hard.”

Meanwhile, Europe has been at the forefront of AI regulations so far.

“China has got its own form of regulatory pathway they’re trying to create.”

He said that the US may most probably be following AI regulations similar to that of the EU.

“There can’t be two sets of regulations. It will be a logistical nightmare for tech companies as it is difficult to follow two separate standards. There will be pressure on the US by big tech companies as it would cost too much for them if there’s a significant difference in policies.

“We’re going through interesting times because generative AI has the potential to be catastrophic. And it will be interesting to see what kind of regulation happens either before or after an incident happens,” he said.

Khan said regulation is about striking the right balance between driving innovation and hindering growth.

AI derives its knowledge from the language which is widely present on the internet. Therefore, it has an inherent bias against languages that are not very much common on the internet.

Therefore, Khan said that there are pitfalls associated with AI that need to be addressed.

Copyright Business Recorder, 2023