Policymakers face a significant challenge in regulating artificial intelligence (AI) as innovations quickly expand while citizens demand action to ensure AI applications don’t endanger their safety, according to Dr Kevin Hebner, managing director, global investment strategist with Epoch Investment Partners.
Given the market imperative, big technology companies have turbocharged their AI efforts to ensure they are on the front of the wave, more interested in speed than safety. Given the economic imperative to be first, it is not surprising the public wants increased transparency and is supporting increased regulation. However, regulators should not act too quickly to clamp down on AI, according to Dr Hebner.
“The track record of regulation suggests one major risk is a rush to action, without the benefit of rigorous cost-benefit analysis and a firm understanding of how the technology is evolving. As often occurs, regulators would inflict a lot of harm in their vain attempt to do a little good,” Dr Hebner said in a new whitepaper, AI: How to Regulate an Emerging Tech?
“A second risk is strangling innovation, as frequently transpires in Europe, while a third is regulatory capture, which seems likely given the high stakes and dearth of AI expertise in government,” he said.
According to Dr Hebner, a balance needs to be struck between encouraging innovation in AI and ensuring safety for citizens. The US is usually much better at this than Europe, which has stifled innovation through regulation.
“This helps explain why most of the top AI professionals are based in the US or Canada even though they were born abroad. It also clarifies why America captures the lion’s share of private sector investment in AI,” he said.
“Mistakes will be made, and they will have important implications for the evolution of AI, the structure of the industry and the cash flow earned by investors. Implementing a rigid and complex regulatory framework is likely to impose excessive costs but do little to protect society. Unfortunately, such an outcome seems highly likely given the political pressure to act, even though we have little idea what the AI ecosystem is going to look like just a few years down the road,” he said.
The whitepaper notes that populations globally are apprehensive about AI and demanding regulation.
“Regarding the economic impact of AI, most people are concerned it could eventually replace their jobs and result in further concentration and power in the technology sector. On a more positive note, many people are optimistic AI will improve the quality of services they receive, especially in healthcare.”
But caution is needed before laws are made to regulate the unknown. “Nobody possesses a crystal ball and we do not know which start-up companies will become the next titans, and which current superstars will fall. This level of uncertainty means we are regulating what we do not really understand, which is, which governments need to be cautious now about introducing restrictive laws.
When it comes to investing in AI for long term returns, Dr Hebner says only a small number of companies will be winners, and will successfully capture the value inherent in AI.
“Nevertheless, we have previously said we view AI as the fourth wave of digital technology after the PC, internet and mobile phones, and this view hasn’t changed.
“We believe AI will be the key driver of equity markets over the next decade, significantly impacting the labour market, productivity and sector concentration, as well as margins and free cash flow generation,” Dr Hebner says.
