The world of artificial intelligence is going through very active and fertile days. It is so fertile that Microsoft, despite laying off 10 thousand employees, can make long-term and billion-dollar investments in artificial intelligence without blinking an eye.
And this rise of machines seems to put a lot of well-paid employees in difficulty.
Pengcheng Shi, Associate Dean of Programming and Information Science at Rochester Institute of Technology, said: “Artificial intelligence will replace white-collar workers. I don’t think anyone can stop it, it’s not a lie, it’s a reality.”
Shi says that many industries such as finance, health, publishing are in danger, but although AI continues its astonishing development, he also says that people can learn how to restrain this technology.
Passing the test and doing the job
Artificial intelligence has already started to stand out in some areas. Especially ChatGPT showed that it could be a candidate for many jobs. When it was revealed that the publisher CNET printed news for artificial intelligence, the reaction they received from social media caused them to take a step back. The academic world was also shaken by the news that ChatGPT scored better than many people in MBA exams in the USA. Darren Hick, Professor of Philosophy at Furman University in South Carolina, said he felt “humiliated” when he realised that a student had cheated using ChatGPT.
Chinmay Hegde, Associate Professor of Computer Science and Electrical Engineering at New York University, said: “Darren Hick and others are right to be concerned. “Certain jobs such as journalism, higher education, graphics and software are under threat of being filled by AI.” Hegde describes the current state of ChatGPT as “very, very good, but not perfect”. At least for now.
Here are the professions that are most vulnerable to fast-learning and ever-evolving technology…
EDUCATION
Shi says that ChatGPT, which is currently banned in New York schools in the US, “can easily teach lessons”. Shi says that the software will be most effective in middle school and high school level courses, and that “although it has errors and informational inaccuracies, these can be easily corrected, in short, it is enough to train ChatGPT for this.”
For higher education, however, both Shi and Hegde believe that courses will continue to need a human leader, at least for the foreseeable future.
FINANCE
In an interview with the New York Post, Shi said Wall Street could also see a lot of layoffs as ChatGPT-like robots continue to develop: “It will affect trading, but in investment banking, you hire university graduates and have them work like robots for two or three years modelling Excel. You can get AI to do that much, much faster.”
Shi is confident that important financial and economic decisions will remain in the hands of humans, even if the data sheets are in the hands of artificial intelligence.
LAWYER AND COMPANY MANAGEMENT
AI can find a place in both courtrooms and boardrooms. ChatGPT passed four of the law course exams at the University of Minnesota at the C+ level. ChatGPT, who also passed the MBA final exam designed for the Wharton School in Pennsylvania, is underlined to be very knowledgeable about the work of analysts, managers and consultants.
DOCTORATE
Artificial intelligence can diagnose a patient in a few seconds.
Dr Prithvi Santana, a recent graduate from the University of New South Wales, gives ChatGPT simple medical information and asks it to diagnose with these. While explaining the result, she says the following: “ChatGPT can do my job as a doctor. The scary thing is that it can process the fine details in the patient history I give to the robot and diagnose the patient for me.”
Dr Santana says that this functionality of ChatGPT excited him so much that his nose started to bleed as a result, but artificial intelligence could not help him in this regard.
SOFTWARE ENGINEERING
Hegde warns that the jobs of website designers and engineers responsible for relatively simple coding are also under threat: “I’m worried about those who do this kind of work. Now that I can ask ChatGPT to build a site for me, it becomes unnecessary to do business with all the people whose routine job it is.”
In essence, AI can design the code needed to make websites based on user requests and parameters, or to be used in other areas of IT.
Shi says that relatively uncomplicated software design jobs will also become obsolete by 2026 or earlier: “As time goes on, maybe today or in the next three, five or ten years, I don’t think there will be much need for software engineers whose job it is to know how to code.”
JOURNALISM
Artificial intelligence technology has made a somewhat tumultuous entry into the newsgathering business: Before CNET’s attempt to print news (and subsequent efforts to edit it), the Guardian also had a GPT software print news in 2020, both with mixed results.
Still, Hedge thinks he’s found a job where AI technology will be more than adequate:
“Proofreading is definitely an area where AI will be very successful. It would certainly do a good job of summarising, making an article concise and so on.”
The only big problem – and for reporters and editors, for now, it’s a salvation – is that the tool can’t fact-check well enough.
Hedge describes it this way: “You can ask a reporter to write an article, to write a story with citations, but often those citations turn out to be made up. This is a known bug of ChatGPT, and to be honest we don’t know how to fix it.”
GRAPHIC DESIGN
OpenAI, the developer of ChatGTP, released a new tool in 2021: DALL-E. DALL-E has the ability to generate images according to the commands given by the user. Along with Craiyon, Stable Diffusion and Midjourney, which do similar work, this tool is a threat to many in the graphic and creative design industry, Hedge says, adding:
“It used to be that you would ask a photographer or graphic designer to create an image for a website. That can now be done with technologies like ChatGPT.”
Shi says he had DALL-E paint a cubist-style portrait of rabbits for the Lunar New Year, and the result was great. “I also wanted a Matisse-style image, but it was not as successful as the Picasso-inspired painting,” he says.
There are also copyright issues with AI-based images. Getty Images has taken legal action against Stable Diffusion’s parent company, Stability AI, claiming that the software “copies and processes millions of copyrighted images”.
fikrikadim/ancient idea editors compiled the news from the New York Post and Mirror.
