Artificial intelligence (AI) performs a great deal of good and will continue to deliver numerous benefits for our contemporary environment, but there will unavoidably be harmful repercussions. We will be better positioned to mitigate and manage the risks the sooner we consider what they might be.
Some famous individuals, such as Tesla and SpaceX CEO and innovator Elon Musk and legendary physicist Stephen Hawking, believe AI might be highly hazardous; Musk once compared AI to the threats posed by the North Korean dictator. Bill Gates, the co-founder of Microsoft, agrees that there are reasons to be cautious but that the positive can exceed the bad if adequately controlled. The moment has come to assess the hazards posed by artificial intelligence, as recent advancements have made super-intelligent computers attainable far sooner than initially anticipated.
With all the buzz surrounding Artificial Intelligence – self-driving cars, robots, etc. – it’s tempting to imagine that AI does not affect our daily lives. In truth, most of us encounter Artificial Intelligence in one form or another nearly daily. From the moment when you wake up and check your smartphone to the time you watch another Netflix-recommended film, AI has rapidly become a part of our daily life. According to a study by Statista, the worldwide AI market is projected to expand by up to 54 percent annually. But what is AI exactly? Will it genuinely benefit humanity in the future? In this post, we will analyze the many pros and cons of Artificial Intelligence.
What is Artificial Intelligence?
From SIRI to autonomous vehicles, artificial intelligence (AI) is advancing swiftly. AI encompasses everything from Google search algorithms to IBM’s Watson to autonomous weapons. In science fiction, AI is commonly portrayed as robots with human-like traits.
Today’s artificial intelligence is classified as narrow AI (or weak AI) since it is designed to execute a particular purpose. However, many researchers’ long-term objective is to develop generic AI (AGI or strong AI). While narrow AI may surpass humans in some tasks, like playing chess or mathematical calculations, AGI would exceed humans in most cognitive tasks.
Is Artificial Intelligence a Threat?
As AI becomes more sophisticated and pervasive, so are the voices warning of its current and potential perils. Concern abounds on various fronts, including the rising automation of some jobs, gender and racial bias issues stemming from outmoded information sources, and autonomous weapons that function without human oversight. And we’re still very early in the process.
Destructive Superintelligence, also known as artificial general intelligence generated by humans that escapes our control and wreaks havoc, is unique. It is also something that may or may not occur (theories vary), so at this point, it is less of a risk and more of a hypothetical threat — a constant source of existential dread.
Here are several ways in which artificial intelligence poses a grave threat:
1. AI Bias
Since people create Artificial Intelligence algorithms, they may contain biases introduced either purposefully or unintentionally by their creators. If AI algorithms are constructed with a preference, or if the data in their training sets are biased, they will deliver limited results. This reality could result in unforeseen repercussions similar to those observed with biased recruiting algorithms and Microsoft’s racist Twitter chatbot. As organizations develop AI algorithms, they must be ethically designed and trained.
2. Autonomous Weapons
AI programmed to perform something dangerous, such as autonomous weaponry designed to kill, is one way AI might pose a threat. It is possible to anticipate that the nuclear arms race will be replaced by a global arms race, including autonomous weapons. Vladimir Putin, the president of Russia, stated, “Artificial intelligence is the future, not only for Russia but for all of humanity. It presents great opportunities and hazards that are impossible to forecast. Whoever becomes the leader in this field will become the world’s ruler.”
A more pressing worry than the possibility that autonomous weapons could develop a “mind of their own” is the potential threat they pose when used by an individual or nation that does not respect human life. They will likely be difficult to dismantle or combat once deployed.
3. Security, Privacy, and the Emerging Deepfakes
Even though job loss is now the most serious risk associated with AI disruption, it is only one of several potential dangers. In a paper titled “The Malicious Use of Artificial Intelligence: Forecasting, Preventative measures, and Mitigation,” published in February 2018, 26 researchers from 14 institutions (academic, civil, and industry) enumerated several other threats that could cause severe harm or, at the very least, sow minor chaos in less than five years.
“Malicious use of AI,” they wrote in their 100-page report, “could threaten digital security (e.g., through cyber-criminals are training AI-enabled machines to try to attack or socially engineer victims at superhuman levels of performance), physical security (e.g., non-state actors weaponizing consumer drones), and political security.”
In addition to its existential threat, Ford is concerned about how AI will negatively impact privacy and security. He cited China’s “Orwellian” usage of facial recognition technology in offices, schools, and other locations as a prime example. However, that is only one country—a “global ecosphere” of enterprises specializing in comparable technologies and selling them worldwide.
We can only speculate whether this technology will ever become standard practice. As with the internet, where we carelessly sacrifice our digital data on the altar of convenience, would round-the-clock, AI-analyzed monitoring one day appear to be a reasonable trade-off for more excellent safety and security despite its nefarious use by bad actors?
Ford stated, “Authoritarian regimes use or will utilize it.” “The question is, to what extent does it infiltrate Western countries and democracies, and what restrictions do we impose?”
4. AI Terrorism
Similarly, there may be new AI-enabled forms of terrorism to contend with, such as the growth of autonomous drones, the introduction of robotic swarms, remote attacks, and the transmission of disease via Nano-robots. Our law enforcement and military forces will need to adapt to their possible threat.
It will require time and substantial human reasoning to decide the optimal approach to prepare for a future that includes even more AI applications so that the potential for negative consequences associated with its continued acceptance can be mitigated to the greatest extent possible.
As with any disruptive event, these are not simple circumstances to resolve. However, as long as humans are involved in establishing solutions, we will be able to reap the benefits of artificial intelligence while minimizing and mitigating its harmful effects.
A robot is one application of AI displacing jobs and creating unemployment (in a few cases). Consequently, others argue that unemployment is always possible due to chatbots and robots replacing humans.
In some technologically advanced nations, such as Japan, robots are regularly used to replace human labor in the manufacturing sector. However, this is not always the case, as automation generates extra employment opportunities for humans while also replacing humans to boost efficiency. AI technologies automate the majority of mundane and repetitive operations. Since we no longer need to memorize information or solve puzzles to complete our jobs, we use our brains less and less. This reliance on AI could pose concerns for future generations.
6. Privacy intrusion and social ranking
It is currently possible to track and evaluate an individual’s every online and offline action. There are cameras practically everywhere, and facial recognition algorithms can identify you. This information will be used to power China’s social credit system, which is projected to assign every one of the country’s 1.4 billion inhabitants a personal score based on their behavior, such as whether they jaywalk, smoke in non-smoking places, or spend too much time playing video games. When Big Brother is watching you and making judgments based on that information, it is not only a violation of your privacy, but May soon led to social oppression.
7. Stock Market Instability
Have you ever wondered how algorithms could bring our entire banking system down? Specifically, Wall Street. You may want to pay attention. It is possible that algorithmic trading will trigger the next major financial catastrophe.
How is algorithmic trading defined? This sort of trading occurs when a computer executes trades based on pre-programmed instructions without the instincts or emotions that could impair a human’s judgment. These computers can execute high-volume, high-frequency, and high-value trades, resulting in significant losses and considerable market volatility—algorithmic HFT (High-Frequency Trading) is proven to be a substantial market risk element. HFT is essentially when a machine executes hundreds of trades at breakneck speed, intending to sell them a few seconds later for modest profits. Thousands of these transactions every second can amount to a substantial sum of money. The problem with high-frequency trading is that it disregards the interconnections of the markets and that human emotion and reasoning continue to play a significant role in our needs.
A sell-off of shares outstanding in the airline market may frighten people into selling their stakes in the hotel industry, which might snowball into people selling their interests in other travel-related companies, which could then affect logistics and food supply firms, etc.
Consider the “Flash Crash” of May 2010, for instance. The Dow Jones fell 1,000 points (more than $1 trillion in value) near the end of the trading day before recovering to normal levels 36 minutes later. What led to this collision? A London-based trader named Navin Singh Sarao initially precipitated the disaster, which was then worsened by high-frequency trading (HFT) computers. Sarao utilized a “spoofing” algorithm to place an order for thousands of stock index futures contracts with the expectation that the market would decline. Instead of completing the wager, Sarao planned to change his mind at the last minute and purchase the stocks that were being sold off at a discount due to his wager. Other people and HFT computers interpreted this $200 million wager as a hint that the market would crash. In response, HFT computers initiated one of the most significant stock sell-offs in history, resulting in a temporary worldwide loss of over $1 trillion.
Financial HFT algorithms are also not always accurate. We perceive computers as the pinnacle of correctness, yet AI is just as intelligent as the humans who designed it. In 2012, a problem brought Knight Capital Group close to insolvency. Computers at Knight inadvertently transmitted numerous orders per second to the NYSE, causing massive havoc for the company. In under 45 minutes, the HFT algorithms performed a staggering 4 million deals involving 397 million shares. Due to the instability caused by this computer glitch, Knight lost $460 million overnight and was forced to be acquired by another company. Knight learned this lesson the hard way, as errant algorithms have enormous ramifications for investors and the markets themselves.
The Bottom Line
Now that you know both the benefits and drawbacks of Artificial Intelligence, one thing is sure: AI has enormous potential to make the world a better place to live. Humans’ most crucial responsibility will be to ensure that the rise of AI does not spiral out of control. Although the merits and disadvantages of artificial intelligence are arguable, its impact on the global economy is evident. It continues to expand daily, contributing to the sustainability of businesses. AI literacy and skill enhancement are unavoidable requirements to succeed in several vocations in the 21st century. The AI and Machine Learning certification course, AI program, and Master in Artificial Intelligence offered by Simplilearn will accelerate your career in AI and prepare you for one of the most exciting careers in the world. This curriculum covers AI fundamentals and advanced topics, including deep learning networks, natural language processing, and reinforcement learning. Start today with this training and establish your AI career aspirations.
Source@techsaa: Read more at: Technology Week Blog