Artificial Intelligence (AI) is a disruptive technology that offers the promise to usher in the next industrial revolution. However, despite the boom in AI applications, there are deepening concerns that the technology may negatively impact jobs, national security, privacy and spread misinformation.
AI is the next evolution in machine learning. AI technologies enable computers to perform a variety of advanced functions, including the ability to understand and translate written language. It is capable of creating new content such as text, images or audio as well as analyzing vast amounts of data.
That's a laymen's definition of the technology. AI is a board field that encompasses many different disciplines, including computer science, data analytics and statistics, hardware and software engineering, linguistics, neuroscience and even philosophy and psychology.
AI burst into the American consciousness with the release of ChatGPT, free app developed by OpenAI, an AI and research company. The app facilitates an almost human-like conversation with a chatbox that answers questions and can assist with tasks such as composing emails, essays or even creating poetry.
Released in November of last year, CatGPT is the fastest-growing app in history, garnering more than 100 million active users. Nearly every teenager has downloaded the app on his or her wireless phone. This writer has been experimenting with the app for months. Think Apple's Siri on steroids.
One measure of the AI boom is the stock price of firms operating in the AI space, including tech giants Microsoft, Goggle and Open AI. No firm has benefited more than Nvidia, which has ridden the wave to a 222% increase in market value. Nvidia makes a powerful chip that's become the workhorse for AI.
AI has attracted the interest of Congress and a legion of critics and champions. Yet there is no denying there are many innovative applications for AI to automate workflow and processes, reduce human errors and eliminate repetitive tasks.
AI is being deployed in the healthcare industry at a dizzying rate. Since health cost are nearly one-fifth (19.7%) of the total U.S. economy, the potential value of AI in healthcare from the administrative side to the delivery of healthcare has enormous potential to reduce costs and improve efficiency.
The technology is already being used by the Centers for Disease Control to analyze public health data. Increasingly, AI is being deployed to assist in analyzing imaging data from MRI's and CT cans. AI can handle some tasks preformed by radiologists, a profession in the throes of declining specialists.
In one example, AI is being used to analyze cell images to determine which drugs are most effective for patients with neurodegenerative diseases. Conventional computers are too slow to spot changes in neurons when patients are treated with different drugs.
That's just for starters. AI is being deployed in medical training, to assist medical professionals in clinical settings, remote monitoring of patients and for diagnostics. One AI software can detect current issues and predict the patient's likelihood of developing the breast cancer in the next several years.
Beyond healthcare, virtually every industry is looking at ways to incorporate AI into their business. Microsoft and Google are working to integrate cutting edge AI into their search engines. Engineering firms are finding ways AI can make their teams more effective.
The software industry is exploring ways to use AI can eliminate the need for certain tasks generally performed by early career or junior programers. Digital news platforms are employing AI to create stories, edited by news people. Schools are eyeing ChatGPT as a tool for students and teachers.
There is no question AI potentially will replace some workers, especially with accelerated advancements in the technology and industry's rapid integration of AI into their businesses.
Another issue virtually unreported is the energy and resource drain that will be created with the growth of AI. A report rom the School of Engineering and Applied Science at University of Pennsylvania raises concerns as AI applications begin to scale up exponentially.
An estimate from the Semiconductor Research Corporation predicts the increasing deployment of AI will "soon hit a wall where our silicon supply chains won't be able to keep up with the amount of data created." Computer memory is stored on components made from silicon.
Companies operating AI systems store data in massive facilities all over the country. These facilities carbon emissions doubled between 2017 and 2020. These centers consume on the order of 20 to 40 megawatts of power, roughly enough to power 16,000 households with electricity.
Like many technologies, there is a dark side to AI. Google's CEO Sundar Pichai is among a growing number of business leaders flagging the capability of the technology to fabricate imagines of public figures and average Americans that are nearly indistinguishable from reality.
Imagine in a political election AI is used to produce a video and audio fake of a candidate making racist or anti-American statements. What if the forgery goes viral on social media before it can be detected? These so-called deepfakes have nabbed the attention of the Department of Defense.
In the context of national security, a fake could dupe military or intelligence personnel into divulging sensitive information to an adversary, posing as a trusted colleague. The Pentagon recently awarded a contract to a startup DeepMedia to design a deepfake detection computer.
Many in Congress have been calling for guardrails to regulate AI. Before Congress could act, the White House announced that seven of the nation's top AI developers agreed to guidelines aimed at ensuring the "safe" deployment of AI.
Amazon, Google, Meta, Microsoft, OpenAI, Anthropic and Inflection agreed to the outline. The guidelines are voluntary and there are no penalties for violating the open-ended agreement.
Regulation often can restrain new technologies in the U.S., while foreign competitors are unleashed to push forward and leapfrog American companies. However, in this instance, there needs to be some rules to prevent the misuse of a relatively new technology by a few bad actors.
Once AI becomes embedded in every business, it will be too late to govern the technology's applications without major business and political upheaval.
No comments:
Post a Comment