Half of all work tasks will be handled by machines by 2025 in a shift likely to worsen inequality, a World Economic Forum report has forecast.

The think tank said a "robot revolution" would create 97 million jobs worldwide but destroy almost as many, leaving some communities at risk. Routine or manual jobs in administration and data processing were most at threat of automation, WEF said. But it said there would be a need for new jobs in care, big data and the green economy.

The WEF surveyed 300 of the world's biggest companies, with more than 8 million employees globally. More than 50% of employers surveyed said they expected to speed up the automation of some roles in their companies, while 43% felt they were likely to cut jobs due to technology.

WEF said the pandemic （疫情） had sped up the adoption of new technologies as firms looked to cut costs and adopt new ways of working. But it warned workers now faced a double threat from "accelerating automation and the economic impact from the Covid-19".

"These things have deepened existing inequalities across labor markets and removed gains in employment made since the global financial crisis in 2007-2008," said Saadia Zahidi, managing director at WEF. "The window of opportunity for active management of this change is closing fast."

WEF said currently around a third of all work tasks were handled by machines, with humans doing the rest, but by 2025 the balance would shift. Roles that relied on human skills such as advising, decision-making, reasoning, communicating and interacting would rise in demand. There would also be a sudden large increase in demand for workers to fill green economy jobs, and new roles in areas like engineering and cloud computing.

But it said millions of routine or manual jobs would be displaced by technology, affecting the lowest paid, lowest skilled workers the most. It said millions would need to be re-skilled to cope with the change, while governments would have to provide "stronger safety nets" for displaced workers.

When it comes to work, workers, and jobs, much of the concern of the modern age boils down to the fear that we’re witnessing the final stage of the game, and that there will be nowhere for humans to withdraw as machines take over the last few tasks. The most recent example comes from the use of artificial intelligence (AI). Is there no area of human experience that can’t be replaced by AI? And if not, what is left for humans to do except the tasks involved in taking care of the machines?

At the heart of this concern is our desire for good jobs — jobs that make the most of workers’ natural abilities and where the work provides the worker with motivation and work-life balance. More importantly, good jobs support workers in learning by doing — and, in so doing, deliver benefits on three levels: to the worker, who gains in personal development and job satisfaction; to the organization, which reforms as staff find new problems to solve and opportunities to pursue; and to the community as a whole, which harvests the economic benefits of hosting positive organizations and workers. This is what makes good jobs productive for the organization, as well as engaging and fulfilling for the worker.

Does the ongoing advance of AI threaten to get rid of all the learning, creativity, and meaning that make a job a good job? Certainly, some have blamed technology for just such an outcome. Headlines today often express concern over technological innovation resulting in bad jobs for humans, or even the complete disappearance of certain professions. Some fear that further technology advancement in the workplace will result in jobs where employees are being asked to work in split times or for longer periods over more days.

The problem here isn’t the technology; rather, it’s the way the technology is used — and, more than that, the way people think about using it.

Imagine you’re standing in line to buy an after-school snack at a store. You step up to the counter and the cashier scans your food. Next, you have to pay. But instead of scanning a QR code (二维码) with your smartphone, you just hold out your hand so the cashier can scan your fingerprint. Or, a camera scans your face, your eyes or even your ear.

Now, this type of technology might not be far away. As technology companies move away from the traditional password, biometric(生物识别的) security, which includes fingerprint, face and voice ID, is becoming increasingly popular.

In 2013, Apple introduced the iPhone 5s, one of the first smartphones with a fingerprint scanner. Since then, using one’s fingerprint to unlock a phone and make mobile payments has become commonplace, bringing convenience to our lives. And since last year, Samsung has featured eye-scanning technology in its top smartphones, while Apple’s new iPhone X can even scan a user’s face. But despite its popularity, experts warn that biometrics might not be as secure as we’d imagined. “Biometrics are ideally good, but in practice, not so much” said John Michener, a biometrics expert.

When introducing the new iPhone’s Face ID feature, Phil Schiller, Apple’s senior vice president, said, “The chance that a random person in the population could look at your iPhone X and unlock it with their face is about one in a million.” But it’s already been done. In a video posted on community website Reddi, two brothers showed how they were each able to unlock the same iPhone X using their own face. And they aren’t even twins.

“We may expect too much from biometrics,” Anil Jain, a computer science professor at Michigan State University, told CBS News. “No security systems are perfect.”

Earlier this year, Jain found a way to trick biometric security. Using a printed copy of a thumbprint, she was able to unlock a dead person’s smartphone for police, according to tech website Splinter. “It’s good to see biometrics being used more,” Jain told CBS News, “because it adds another factors for security. But using multiple security measures is the best defense.”