Are We Being Too Naive with AI Technology?

Jan 5, 2022

Welcome to the new year. While we were all dealing with friends and family worldwide that caught the highly contagious Omicron variant, one technology story caught my eye. Perhaps you noticed it too. Alexa, the voice-enabled assistant, offered a potentially deadly challenge to a young 10-year-old.

Kristin Livdahl’s daughter asked Alexa for a challenge. The following is what Alexa replied, "Here's something I found on the web. According to ourcommunitynow.com: The challenge is simple: plug in a phone charger about halfway into a wall outlet, then touch a penny to the exposed prongs."

While that's unsettling, to say the least, one of the essential ideas you might want to take away from this article is the fact that artificial intelligence (AI) is unpredictable. It is not human. It is not moral. And it only functions to complete the objectives for which it was created, so buckle-up for the ride.

As we immerse ourselves fully in the digital era and more and more of our lives become digitized, artificial intelligence is an inevitable topic of discussion. Sure, AI is one of the most exciting technological developments in recent years, and it promises to change how we live and work. It already has.

The concern is that while AI generates many benefits for human life, there may also be risks associated with it. There are ethical risks like privacy breaches or bias in decision-making, economic risks like job loss or automation, and existential risks like technology making unpredictable decisions that humans wouldn't make.

What is Artificial Intelligence?

Artificial intelligence is a broad term describing any machine-learning system designed to behave intelligently. It includes natural language processing systems, speech recognition, computer vision, and more.

In fact, we owe AI our collective thanks (although I would never thank it since it’s not a living and breathing being) because it allowed researchers and scientists to create the vaccines billions injected into their arms in record speed. What typically took about four years took less than one year.

The Benefits of Artificial Intelligence

The reality is that AI can help us in many ways, like automating tasks. By using AI, we won't need to do certain things manually. As a result, it's a massive benefit for businesses and individuals because it does free up time.

In addition to freeing up our time, AI can also help provide faster service or produce better quality work. Imagine there's a machine that, for example, evaluates billions and billions of data sets in a short time, cutting the period for developing life-saving medicines. Yes, as AI did for the coronavirus pandemic.

Additionally, AI can take on some jobs humans aren't good at doing, like recognizing patterns in vast amounts of data quickly and accurately. These are jobs that people find challenging, but machines don't have this challenge thanks to their ability to process information quickly and accurately.

However, there are some risks associated with artificial intelligence that we should address before moving forward with it fully:

The Risks of Artificial Intelligence

The risks of AI are many and varied. A primary ethical risk is that of privacy breaches, where AI can collect and store all kinds of sensitive data about people without their explicit knowledge or consent.

AI can also make mistakes in decision-making, leading to biased outcomes that don't meet the needs of different groups or may even harm them. For example, we've seen that computer algorithms can be trained on datasets that exclude certain races and genders and then generate sorting algorithms based on the same bias.

Finally, some technologists fear that AI will become so advanced and intelligent that it becomes uncontrollable and unpredictable. Just as nuclear power plants have the potential for a meltdown if not run properly, so too do some worry that with improvements in technology come new existential risks. It includes unpredictable decisions that can impact a life, as in the case of Alexa suggesting to a 10-year-old girl that she, essentially, kill herself, but also society as a whole.

1. Privacy Issues

One of the most significant risks with AI is its potential to invade our privacy. As AI becomes more and more advanced, it will be able to process and analyze more information about us than ever before, which will allow for a broader understanding of who we are and what we do. Because AI will have access to a great deal of personal data, the issue of data security becomes critical. Companies need to protect their customers' sensitive information from hackers and other malicious agents.

Another issue is that AI could make decisions that violate our fundamental human rights. Imagine if an autonomous vehicle made a decision not to stop to save five people from being killed by a drunk driver because it deemed one person's life less valuable than five others. In addition, AI is shown to be vulnerable to bias when making decisions involving race, gender, or sexuality. For example, ethicists are sounding the alarm about how AI reinforces gender stereotypes with our bots and voice assistants. Brookings reported that Josie Young, an AI ethicist said, “when we add a human name, face, or voice [to technology]… it reflects the biases in the viewpoints of the teams that built it.”

2. Bias in Decision-Making

Artificial intelligence can make decisions that humans would not be able to make. However, the problem is that artificial intelligence systems also can be biased in how they make decisions. If AI systems could process data and information without bias, this would greatly benefit humanity, but we're not there yet, although many groups are trying to shape AI development. Still, it's essential to acknowledge that the benefits of AI may not come without risk.

There is a possibility of bias when an algorithm gets written by people who have biases themselves. We also have to consider biases in the data inputted into an AI system and biases in the training data gathered from past discriminatory practices. For example, if you trained your AI system on past hiring data, it might learn to discriminate against certain groups of people. If you're using a bias in your training data, your AI system will have the capacity for bias.

3. Technology Making Unpredictable Decisions That Humans Wouldn't Make

One risk of AI is that it can make unpredictable decisions. Technology will always have limits to what it's capable of, but the more advanced the technology becomes, the less predictable the outcomes. There's also the possibility that AI could make decisions outside of human control. While some people believe this is unlikely because humans would need to create the code for an AI system, others say it is possible and has already happened in some cases.

Businesses must thoughtfully consider how to integrate AI into their operations, but people also have to inform themselves on the ethical impact of AI in society as informed citizens. Even someone who is not a technology expert can figure out with enough reading and research that AI is unpredictable and could make decisions outside of human control. Therefore, we must consider these risks before developing or implementing AI-based tools and technologies that affect our lives and future generations.

Artificial Intelligence is a Complex and Complicated Subject

There are many pros and cons to AI, and we're just scratching the surface of what it can do. The topic is so vast and detailed that one post can't cover everything. However, business leaders, workers, and people need to get informed about new technology because of its power.