Does Software that Explains Itself Really Help?
2022-04-09
LRC
TXT
大字
小字
滚动
全页
1Scientists who make artificial intelligence (AI) systems say they have no problem designing ones that make good predictions for business decisions.
2But they are finding that the AI may need to explain itself through another algorithm to make such tools effective for the people who use them.
3AI is an area of computer science which aims to give machines abilities that seem like human intelligence.
4"Explainable AI," or XAI, is a new field that has received a lot of investment.
5Small, new companies and large technology companies are competing to make complex software more understandable.
6Government officials in the United States and European Union also want to make sure machines' decision-making is fair and understandable.
7Experts say that AI technology can sometimes increase unfair opinions about race, gender and culture in society.
8Some AI scientists think explanations are an important way to deal with that.
9Over the last two years, U.S. government agencies including the Federal Trade Commission have warned that AI, which is not explainable, could be investigated.
10The European Union could also pass the Artificial Intelligence Act next year.
11That law would require explanations of AI results.
12Supporters of explainable AI say it has helped increase the effectiveness of AI's use in fields like healthcare and sales.
13For example, Microsoft's LinkedIn professional networking service earned 8 percent more money after giving its sales team AI software.
14The software aims to predict the risk of a person canceling a subscription.
15But the software also provides an explanation of why it makes a prediction.
16The system was launched last July.
17It is expected to be described on LinkedIn's website.
18But critics say explanations of AI predictions are not trustworthy.
19They say the AI technology to explain the machines' results is not good enough.
20Developers of explainable AI say that each step in the process should be improved.
21These steps include analyzing predictions, creating explanations, confirming them and making them helpful for users.
22But after two years, LinkedIn said its technology has already created value.
23It said the proof is the 8 percent increase in money from subscription sales during the current financial year.
24Before the AI software, LinkedIn salespeople used their own abilities.
25Now, the AI quickly does research and analysis.
26Called CrystalCandle by LinkedIn, it identifies actions and helps salespeople sell subscriptions and other services.
27LinkedIn said the explanation-based service has extended to more than 5,000 sales employees.
28It includes finding new workers, advertising, marketing and educational offerings.
29"It has helped experienced salespeople by arming them with specific insights," said Parvez Ahammad.
30He is LinkedIn's director of machine learning and head of data science applied research.
31But some AI experts question whether explanations are needed.
32They say explanations could even do harm, creating a false idea of security in AI.
33Researchers say they could also create design changes that are less useful.
34But LinkedIn said an algorithm's strength cannot be understood without understanding its "thinking."
35LinkedIn also said that tools like its CrystalCandle could help AI users in other fields.
36Doctors could learn why AI predicts that someone is more at risk of a disease.
37People could be told why AI recommended that they be denied a credit card.
38Been Kim is an AI researcher at Google.
39She hopes that explanations show whether a system presents ideas and values people want to support.
40She said explanations can create a kind of discussion between machines and humans.
41"If we truly want to enable human-machine collaboration, we need that," Kim said.
42I'm Dan Novak.
1Scientists who make artificial intelligence (AI) systems say they have no problem designing ones that make good predictions for business decisions. But they are finding that the AI may need to explain itself through another algorithm to make such tools effective for the people who use them. 2AI is an area of computer science which aims to give machines abilities that seem like human intelligence. 3"Explainable AI," or XAI, is a new field that has received a lot of investment. Small, new companies and large technology companies are competing to make complex software more understandable. Government officials in the United States and European Union also want to make sure machines' decision-making is fair and understandable. 4Experts say that AI technology can sometimes increase unfair opinions about race, gender and culture in society. Some AI scientists think explanations are an important way to deal with that. 5Over the last two years, U.S. government agencies including the Federal Trade Commission have warned that AI, which is not explainable, could be investigated. The European Union could also pass the Artificial Intelligence Act next year. That law would require explanations of AI results. 6Supporters of explainable AI say it has helped increase the effectiveness of AI's use in fields like healthcare and sales. 7For example, Microsoft's LinkedIn professional networking service earned 8 percent more money after giving its sales team AI software. The software aims to predict the risk of a person canceling a subscription. But the software also provides an explanation of why it makes a prediction. 8The system was launched last July. It is expected to be described on LinkedIn's website. 9But critics say explanations of AI predictions are not trustworthy. They say the AI technology to explain the machines' results is not good enough. 10Developers of explainable AI say that each step in the process should be improved. These steps include analyzing predictions, creating explanations, confirming them and making them helpful for users. 11But after two years, LinkedIn said its technology has already created value. It said the proof is the 8 percent increase in money from subscription sales during the current financial year. 12Before the AI software, LinkedIn salespeople used their own abilities. Now, the AI quickly does research and analysis. Called CrystalCandle by LinkedIn, it identifies actions and helps salespeople sell subscriptions and other services. 13LinkedIn said the explanation-based service has extended to more than 5,000 sales employees. It includes finding new workers, advertising, marketing and educational offerings. 14"It has helped experienced salespeople by arming them with specific insights," said Parvez Ahammad. He is LinkedIn's director of machine learning and head of data science applied research. 15But some AI experts question whether explanations are needed. They say explanations could even do harm, creating a false idea of security in AI. Researchers say they could also create design changes that are less useful. 16But LinkedIn said an algorithm's strength cannot be understood without understanding its "thinking." 17LinkedIn also said that tools like its CrystalCandle could help AI users in other fields. Doctors could learn why AI predicts that someone is more at risk of a disease. People could be told why AI recommended that they be denied a credit card. 18Been Kim is an AI researcher at Google. She hopes that explanations show whether a system presents ideas and values people want to support. She said explanations can create a kind of discussion between machines and humans. 19"If we truly want to enable human-machine collaboration, we need that," Kim said. 20I'm Dan Novak. 21Dan Novak adapted this story for VOA Learning English based on reporting from Reuters. 22_______________________________________________________________ 23Words in This Story 24algorithm - n. a set of steps that are followed in order to solve a mathematical problem or to complete a computer process 25subscription - n. an agreement that you make with a company to get a publication or service regularly and that you pay for regularly 26gender - n. the state of being male or female 27analyze -v. to study something closely and carefully; to learn the nature and relationship of the parts of something by a close and careful examination 28insight - n. an understanding of the true nature of something 29collaboration - n. to work with another person or group in order to gain or do something