Concerns about AI range from how it will change the nature of work to what it could mean for human life on earth. Credit: Qilai Shen/Bloomberg
As AI continues to advance at a rapid pace, tech moguls and influencers around the world are sharing their views on the expected implications.
With the development comes a whole host of questions that Telegraph readers have sent in in abundance.
Many are concerned about how artificial intelligence will change the nature of their work and whether it will make some jobs redundant. Others wonder how society will change and what the advancement of AI could mean for human life on Earth.
Technology editor for The Telegraph, James Titcomb, answered readers' most pressing questions about artificial intelligence and its future.
Read on for our expert to answer your questions about AI, and please share any additional questions in the comments section below.
Reader Gary Bowers asks: “ How fast is AI evolving? For example, does he study how human emotions work?”
Technology Editor James Titcomb of The Telegraph replies:
“Artificial intelligence is clearly making rapid strides in some areas. , but remains fairly simple in others. One of the problems with the term «artificial intelligence» is that it forces us to think of these systems from a human perspective.
“The reality is that in some areas — for example, in arithmetic — machines surpassed us by decades; in others they are not even close. Emotions are related to the second: we have not created AI that would show anything close to emotions, partly because we do not fully understand how they work in living beings.
“Undoubtedly, there are artificial intelligences that scan faces and voices to detect emotions that are used by the police and marketing companies, although their effectiveness is questioned.
“At this point, it’s probably better to think of AI in terms of individual capabilities rather than comparing it to the human brain. In some areas, such as image creation or text summarization, it is improving rapidly, although these changes tend to come in leaps rather than increments.”
Another reader, Ben Francis. wants to know: «How can we prevent this technology from reaching escape velocity and leaving us behind?»
James replies:
» At the moment, AI can't do much of what we can do, so it's unlikely we'll be left behind anytime soon, but experts have some ideas on how to manage its growth.
“First, we must prohibit AI from writing computer code to develop AI. This will prevent a phenomenon known as «recursive self-improvement», where the system constantly improves itself until it outsmarts people and then becomes all-powerful.
“Another emerging area of research is known as AI alignment: ensuring that a robot’s goals are aligned with ours. It is hoped that this will prevent the famous end-of-the-world scenarios where the AI is given the simple task of cleaning up the oceans or making paper clips, and eventually leading to the destruction of humanity as a side effect.»
0605 AI Impact Roy Reed questions: «Are we our own worst enemies when it comes to AI?»
This is what James says:
< p >“If you tell researchers about concerns about artificial intelligence, or suggest that progress may be slower, many will argue that concessions are pointless. They say that sooner or later someone will develop this technology, and isn't it better for us than for China or Russia?
“As with most technologies, development is often difficult to stop. The only exception was nuclear weapons, whose deployment was halted by a concerted international effort decades ago. Security advocates would like to see similar treaties and international agreements when it comes to AI: some have suggested creating a body similar to the International Atomic Energy Agency to regulate the use of this technology.”
Frédéric Otobele asks : «How will AI affect the work of cybersecurity experts?»
Our expert replies:
«In short, they will be busy! Security experts have been warning for years that AI can lead to the industrialization of hacker attacks as automated systems identify weaknesses in security networks.
“A growing concern is that AI software that can accurately reproduce human voices and images could bypass security controls such as voice banking or automate fraud so that vulnerable people are tricked into sending money.
“We are not ready yet. see it happen, but it's probably only a matter of time. Cybersecurity experts are likely to be in demand more than ever.”
Kates Ferrier requests: «I don't understand why AI is being implemented. Why are we, as a nation, giving way to job insecurity and payroll opportunities? It's scary.»
James says:
“New technologies — from the loom to the steam engine — have often threatened job losses but have ultimately made us more productive, with more jobs still available. The question with AI is whether it represents an incremental change that could make huge groups of the workforce redundant, not just certain jobs. We're not ready yet.
“The other answer is that it's just hard to curb inventions, at least not without very strict regulation. Employers who could save money by using AI are unlikely to voluntarily choose to ignore it.”
ReaderRichard Burgesswould like to know: “Do we need AI ? ”
Our expert answers:
“We often talk about the downsides of AI, and many of the uses that have emerged over the past few months — homework cheating, hacking, or copyright infringement — seem like things we could do without. Proponents of artificial intelligence say it has some very serious benefits, such as making us more productive at work, helping us discover new drugs, and eventually self-driving cars that could make roads safer.
“ Poor productivity in the UK over the past decade and a half has continually puzzled economists and policy makers: if AI can help solve this puzzle, it will certainly be of benefit.”
Mohamad BeheriQuestions: “Of course, if jobs are at risk, that means it's not good for the economy as less taxes are pumped into the system, shouldn't the government intervene to protect citizens?»
Jamesanswers. :
“In general, new technologies do not mean job losses in the long run, but there is active debate about whether tax-paying jobs should be protected in the short term to minimize disruption to individual lifestyles .< /p>
“One of the proposals, supported by Bill Gates and others, is to tax robots and AI in the same way that we tax humans. In theory, this would level the playing field, giving people fewer benefits.
“In practice, AI is hard to define and even harder to tax. But if millions of people become redundant, governments will have to find a way to adapt. Ideas being discussed include raising taxes on wealth or corporate profits to fund universal basic income.”
AI-proof jobs that pay the highest wages Jane Pourtney: : “< /strong>To what extent have the A level and college essay questions been redesigned to overcome AI assistance?»
Here's what our experts have to say:
«One of the first clear effects of ChatGPT was the cheating epidemic. Thousands of students turned in essays and homework generated by the system, leading to the software being banned from some schools and universities.
“Others have turned to anti-fraud tools designed to verify that something was written by artificial intelligence, although many of them make mistakes, causing students who wrote their own work to be falsely accused of cheating.”
“Teachers, it seems, but gradually adapt, and do not resist. Some move essay writing to the classroom where students cannot use ChatGPT. Others allow students to use the software, but add interviews to show that students understand the subject.
“The latter may prove more useful: like calculators and spell checking, students are likely to continue to use AI at work.”
Another reader, Charlie Jones, asks: “If you were a teenager and soon had to make a choice in favor of university and a future career, what would you look for or avoid in terms of AI? ?»
James replies:
«That's a great question. The instinctive answer is computer science or math: if AI replaces jobs, it's safe to say that at least the people who develop it will be working.
«Any skilled physical job is likely to will be in demand for some time: while software is booming, robots remain a problem Lawyers won't be short of work either, judging by the frequent lawsuits against AI companies from people who say they've been slandered or that their data was stolen.
“However, with a few exceptions, many of today's vacancies probably still exist, just in other forms. AI is a tool that still requires human intervention.”
Question from Joe Jackson: “Should we count tech-based films like Will Smith in I, Robot, a very possible reality within the next 50 years?
James replies:
«Probably no. Hollywood movies like The Terminator are good at entertaining us but bad at telling us about the future of AI. For example, they often endow AI with human qualities — a lust for power — evidence of which we have no evidence of their existence.
“50 years is a long time, and AI will undoubtedly achieve tremendous success in these decades.
“We certainly need to be wary of risks, but the most risky scenarios involving AI are probably due to people misusing them in areas such as weapons systems than typical Hollywood examples of a race of robots enslaving people.
Peter Mitchell asks: «Can AI replace customer service operators?»
Our expert answers:
«That's what's already happening. Go to many websites today and you will find yourself talking to an AI bot, not a human operator. Energy supplier Octopus, for example, says that customers actually prefer to communicate with AI than with staff.
“AI is unlikely to be able to answer all questions for many years, but the number of cases it can handle is likely to gradually increase until it solves most customer service questions.
«At the moment, AI is better at responding to chats and emails than to phone calls, but voice recognition and replication technology means it's changing too, although that may seem supernatural to some people.»
>At last, readerPeo Rakatawould like advice: «How can a finance and banking professional adapt to the new AI environment to avoid layoffs?»
In response, James says:
» It's hard to predict, but the white-collar jobs that are likely to be most at risk from AI are those that involve repetitive or data-intensive tasks: data entry and analysis, regulatory compliance, etc. Those that involve more face-to-face interaction are less likely to be affected.This is likely the case for finance and many other office jobs.
“These changes tend to come relatively gradually, and employers often find new jobs for our employees Computers and the Internet have changed offices a lot, but we still have the same number of people working, although we have fewer typists and secretaries.
Do you have questions about AI? comment section below
Свежие комментарии