How can artificial intelligence be defined?
Artificial intelligence (AI), a really revolutionary advancement in computer science, will become an essential component of all modern software in the coming years and decades. This presents both a risk and an opportunity. AI will enhance both defensive and offensive cyber operations. New cyber attack methods will also be developed to take advantage of some AI technology weaknesses. Finally, the enormous amounts of training data that AI requires will raise their worth and change how we must approach data security. Prudent global governance will be required to ensure that this era-defining technology will result in broadly shared safety and wealth.
1.This is a common instance of an AI failure: the inability to recognise images:
Around 20 years ago, the advancement in image identification, often known as computer vision, launched the triumphal march of deep learning, a group of techniques frequently used to build AI. Prior to moving on to tougher and more challenging challenges, it solved previously intractable tasks like telling cats from canines and vice versa. The idea that computer vision is a reliable, dependable technology that is unlikely to fail has gained widespread acceptance in recent years. 7,500 unedited nature images were acquired by researchers from Berkeley, Chicago, and the University of Washington a year ago, and even the most advanced computer vision algorithms struggled to understand them. Despite their provenance, even the most reliable algorithms
2.Humans now hold the jobs that AI once held:
2015 saw the opening of Japan's first avant-garde Henn-na Hotel to tourists. The front desk, housekeeping staff, porters, and in-room attendants were all robots employed by the hotel. However, the bots immediately racked up customer complaints: they frequently broke down, we weren't able to provide acceptable answers to visitor concerns, and in-room staff members frightened visitors at night by mistaking snoring for a wake command. After years of struggle, the hotel's management company succeeded in getting rid of the final one of its expensive, annoying, and unreliable bots, and replaced them with actual staff. The management announced that it will visit the lab once more to determine whether it can create a new generation of hospitality bots that are more skillful.
3.AI-powered cart A malfunction on the runway:
On the tarmac, a food cart that was AI-driven malfunctioned, circling out of control and getting closer to a gate-parked, vulnerable Aeroplane. The cart was finally brought to a stop by a yellow-vest worker who ran over it with another vehicle and knocked it over. This case has veered off course a bit. The cart was neither mechanically or artificially intelligently controlled in any way. This is regarded as one of the more well-known AI blunders.
4.An AI-related loss resulted in the filing of a lawsuit:
A Hong Kong real estate magnate invested in an AI system to manage some of his money in order to boost resources. The robot actually carried on losing up to $20 million per day. The tycoon sued the fintech service provider for US$23 million in an effort to recover some of his money. The complaint, which asserts that the company exaggerated K1's capabilities, is the first instance of a judicial proceeding involving losses from automated investing. Another illustration is this AI failure.
5.AI to treat cancer may result in patient death:
IBM spent US$62 million on yet another failure in an effort to create an AI system that would help in the fight against cancer. However, the result was once more unsatisfying. A physician at Jupiter Hospital in Florida claimed that the product was an abject failure. He continued by saying that they bought it for marketing purposes. According to medical professionals and consumers, Watson encouraged doctors to give a cancer patient who was experiencing major bleeding a medicine that would make the bleeding worse. Medical professionals and patients alike reported several instances of harmful and incorrect treatment recommendations.
6.AI thinks that Congresspeople are similar to criminals:
Another face recognition mistake was caused by Amazon. Its AI system was designed to identify criminals based on their facial images, but when tested using a collection of images of members of Congress, it not only produced inaccurate results but also exhibited racial discrimination. The ACLU (American Civil Liberties Union) said that despite the fact that people of colour only make up 20% of Congress, approximately 40% of Rekognition's (the system's) incorrect matches in our test were of individuals of colour. It's unclear whether the issue was with non-white face recognition or the training set of data was biassed. Probably both of them. However, merely using AI to assess whether or not a person is a person
7. AI-RELATED WEAKENING OF ETHICS AND GOOD WILL:
Even religious leaders are raising concerns about the potential socio-economic drawbacks of AI, joining technologists, journalists, and political figures in doing so. Pope Francis warned against artificial intelligence's propensity to "circulate tendentious opinions and false data" in a 2019 Vatican summit themed, "The Common Good in the Digital Age," and emphasised the serious repercussions of allowing this technology to advance without proper monitoring or restraint. He said, "This would unfortunately lead to a regression to a form of barbarism dictated by the law of the strongest." If mankind's "so-called technological progress" were to turn into an adversary of the common good, he said. These worries have greater weight thanks to the quick ascent of the conversational AI tool ChatGPT.
8. Enhanced Hacking:
Artificial intelligence accelerates progress and, in many circumstances, surpasses our capacity for comprehension as humans. Automation makes it more difficult for humans to detect evil deeds like phishing, the introduction of viruses into software, and the manipulation of AI systems for personal gain due to the way these systems see the world.
CONCLUSION
On numerous occasions, using AI has a detrimental effect on society. A few AI recruiting systems have been found to be discriminatory towards women. The arrest of innocent men and women was caused by false facial recognition matches. Major traffic accidents have been caused by self-driving cars using flawed AI algorithms, and in certain cases,
Comments
Post a Comment