Artificial Intelligence: Separating the Hype from Reality

Artificial Intelligence: Separating the Hype from Reality
Artificial Intelligence: Separating the Hype from Reality

Like bees for honey, technology trends generate hype. Just adding the word “dotcom” to the name of the company raises stock prices in the days of Internet salads. Cloud computing, big data, and each cryptocurrency have taken a turn in the cycle of sensations in recent years. Every trend brings truly promising technological developments, screwing up keywords, enthusiastic investors, and convincing consultants who offer enlightenment – for a fee, of course.

Now the catchall phrase of artificial intelligence forms as the current technological trend. However, because the claims to be achieved are very large, businesses risk increasing their expectations for A.I. too high – and wasting money trying to apply technology to problems that cannot be solved.

Consider bubbling warning signs. Venture capitalists are eager to fund A.I. They risked 1,028 startups related to A.I last year, up from 291 in 2013, said the PitchBook researcher. Twenty-six of these companies have “A.I.” in their name, compared to the previous five years. Then there are many promising conferences to explain A.I. to a surly manager. At the annual World Economic Forum meeting in Davos, Switzerland, this year’s agenda included no less than 11 panels that referred to A.I., with names such as “Designing A.I. Strategy “and” Establish Rules for A.I. Race. “(Fortune was also involved in this action: 2018 Global Technology Forum in Guangzhou, China, dominated by discussion. A.)

The result is a serious subject running the risk of jumping a shark. “If advocates are not careful, they will succeed in Bitcoinized A.I.,” said Michael Schrage, a researcher at the MIT Initiative on Digital Economics.

Don’t get me wrong – artificial intelligence is more than just fad. This represents a new way of doing business with turbocharging trends in automation, sensor-based industrial monitoring, and algorithmic analysis of business processes. Computer science has helped machines perform routine tasks faster than humans. New techniques A.I. – combined with the ever-increasing computational power and accumulation of years of digital data – means that for the first time computers have learned the tasks humans need instead of just doing what they are told.

The result, said Tom Mitchell, a professor of machine learning at Carnegie Mellon University, was no less than “one of the main forces for society and the lifestyle of thecoming years.” IDC researchers estimate spending on A.I. will approach $ 80 billion in three years. Paul Daugherty, Accenture’s chief technology and innovation consultant, considers that figure will prove to be low because “it doesn’t take into account investment companies that are transforming around A.I.”

However, as with interesting technology, there are limits to what A.I. can reach. Self-driving cars are a perfect example. We already have technology for them to operate in ideal conditions, but even John Krafcik – CEO of the subsidiary of the Alphabet Waymo independent car – acknowledges that they will never be able to drive in all weather conditions without human input. What’s more, computers are very good at learning clear tasks, such as identifying people in photos or copying speech accurately. But understanding human motivation or drawing nuanced conclusions from text – insights into which humans excel – remain out of the reach of machines. Mitchell said from CMU, “We are still in the earliest stages of trying to produce it.”

What is A.I. not yet able to provide comfort for the CEO. Susan Athey, a professor of technology economics at Stanford University, assured managers in valuable executive education courses – as well as limitations on A.I. the scientist they employ. “All new doctors are bought, but they don’t have experience about what doesn’t work, the project can’t do it,” he said. A.I., said Athey, can be justified “feels magical.” But the best is to analyze the situation that has been prepared by the designer to be interpreted, as opposed to making decisions about subjects that have never been seen before. “It’s not right that you will arrange for you,” Athey said.

A.I., in other words, not silver bullets. Jean-François Gagné, CEO of Montreal Element AI’s software startup, reminded clients that A.I. the solution is only as good as the accumulated data entered into it. “The opportunity that is seen by every organization is the ability to have an adaptive system,” he said. “This is a trip. This is not something you can buy and suddenly flips the switch. With the definition of A.I., it takes time to study. ”

Gagné analogues the process of building A.I. useful for the difference between “teaching your children the right thing versus getting the right behavior in adulthood.” It will take at least as long as you know whether the business can understand A.I. this correctly. when – or if it is a very expensive and difficult to understand money pit.


Written by Siva Prasanna

Leave a Reply

Your email address will not be published. Required fields are marked *

cloud computing trends

Top Trends Of Cloud Computing In 2019

Inventory Tracking Software for Small Business 4

Inventory Tracking Software for Small Business