The Risks of AI and How Policy Can Control the Future of the Industry

The effects of artificial intelligence technology on policy have been broadly discussed and evaluated, but what about the effects of policy on AI? And what about the ways in which actors within the AI industry may influence policy?

AI technology has real potential to influence economic growth, increasing productivity, reducing costs and even contributing to the establishment of new products and markets, among other things.

However, the potential positive economic impact of AI also introduce unavoidable potential for negativity too. So, in order to properly realise the potential of AI, it’s necessary for the technology to be monitored and regulated.

So, how is this actually being done? That is, how is the AI industry being monitored and regulated, and how will AI-related policy impact the future of the industry as a whole, and vice versa?

 

The Risks Associated with AI

 

There’s no doubt about the fact that AI technology has massive potential to transform many industries, and in some ways, it already has. But, along with its potential, AI has a lot of risk associated with it that could be detrimental to the industry if it’s not managed and regulated properly.

Here are some of the most fundamental risks of AI.

 

1. Lack of Transparency

 

Due to the complex nature of AI, explanations are often not given for how the technology works, how it does what it does and more. Although the reasons behind this lack of transparency aren’t necessarily nefarious, it does give rise to companies being able to hide things like the dangers of their tools.

 

2. Social Manipulation and Socio-Economic Inequality

 

A big fear associated with AI is the way in which it may be used to enforce social manipulation by means of skewed algorithms on social media. This risk is particularly associated with politicians and political issues.

The other risk is the apparent socio-economic biases that are implicit within AI technology, which results in a significant risk of perpetuating discriminatory hiring processes that are industries are currently attempting to eliminate.

 

3. Job Losses

 

Ever since the recent explosion of AI technology took the world by storm, the most widespread concern has been about potential job losses as a result of increased automation.

 

4. Loss of Privacy

 

While the primary concern of individuals may be potentially losing their jobs, businesses seem to be most worried about data privacy. The increasing use of AI technology to manage private information gives rise to concerns surrounding the ability of the technology to keep it secure, as well as the rules that dictate how this needs to be done.

 

5. Social Surveillance

 

The general public has become increasingly concerned about current and potential social surveillance that may be conducted by means of AI. The proliferation of this fear was largely spurred on by China’s use of facial recognition software and how this may compromise individuals’ safety, privacy and autonomy.

 

6. The Development of Autonomous Weapons

 

Many industry experts have made their fears about the potential of AI-generated autonomous weapons clear, raising the potential for the start of a modern AI global arms race if the technology isn’t controlled now.

 

7. Possible Financial Crisis

 

The use of AI in the financial sector is becoming increasingly popular, especially in everyday finances as well as trading, resulting in concerns associated with algorithmic training and risks surrounding possible failures of the technology.

 

8. Overreliance on AI and Loss of Humanity

 

A general concern across the board is that using AI in so may different ways in a variety of industries will lead to a loss of humanity. In particular, human empathy, creativity and reasoning will be reduced, as well as communication between individuals, especially in business.

 

 

9. AI Becoming Self-Aware

 

Many people, both experts and laymen alike, have been outspoken about their fears of AI technology becoming self-aware. The potential for AI technology to become sentient and develop beyond human control creates huge risks for the future of humanity, regardless of how unlikely this may be.

 

10. Platform for Criminal Activity

 

AI is becoming increasingly accessible, by design, and this creates the possibility for nefarious actors to use it for criminal activity. This includes online predators, scammers and more.

 

How Can Policy Be Used to Manage and Regulate AI?

 

The purpose behind government policy related to artificial intelligence is to manage the industry as a whole and mitigate the risks posed by emerging technology.

Of course, modern companies have their own internal AI policies these days (since it’s a legal requirement in most places), but since a great deal of the capabilities of AI are still unknown, it’s especially important that regulators also keep a close eye on what’s going on within the industry to prevent any serious problems.

The US, for instance, has proposed and passed a series of AI bills recently, most of which have the potential to shape the ways in which the industry will develop in the future. In fact, the US has gone as far as establishing the United States AI Safety Institute, and a bill has been proposed that will not only authorise its existence as a federal body but will also allow it to create standards and guidelines to be enforced nationwide.

A few of the other Senate bills recently proposed pertain to the influence of AI on education and business regulation.

Essentially, these bills and, eventually, laws, will allow world leaders and regulatory bodies to protect individuals, businesses and other groups of people against the possible dangers of AI, and reduce the risks before they become problematic. The idea is also that these regulations will help authorities weed out people and parties who intend on using AI for potentially criminal purposes.

But, what about businesses within the AI industry?

Well, first and foremost, the ways in which they choose to develop and make use of AI will influence the risks that arise in the industry, and a lot of this has to do how they choose to regulate themselves and the ways in which they respond to AI policy.

In an interesting move, OpenAI, a well-known industry leader, has publicly endorsed several US senate bills pertaining to AI safety regulations.

In a statement made by the company’s Vice President of Global Affairs, Ana Makanju, Open AI expressed its commitment to safety in the industry, saying that they “…have consistently supported the mission of the institute, which leads the US government’s efforts to ensure that frontier AI systems are developed and deployed safely.”

This very public endorsement, however, is almost definitely more than just an ethical move, however. Industry experts speculate that OpenAI may simply be trying to get on the right side of policymakers and regulators, so to speak, in anticipation of the fact that the industry giant is likely to face serious scrutiny in the future on a regular basis. The idea is that, essentially, they’re hoping to set the stage for a positive relationship between themselves and the government which is, by no means, A bad thing.

Of course, since we’re talking about the risks posed by AI and how regulators are hoping to mitigate these risks using policy, it’s important to consider how actors within the AI industry may be influencing the formalisation of policy.

Essentially, the influences of policy on AI and AI on policy are multi-directional and complex, and while it’s too soon to tell exactly what the ramifications of these moves will be in the long term, one thing’s for certain: strict policies, regulations and protections pertaining to AI technology are necessary to mitigate risks in the industry and protect both direct players and indirect users.