Finding the Right Balance: Human-AI Combinations for your Business | Technology
 |
| Finding the Right Balance: Human-AI Combinations for your Business |
Choosing the Right Balance: Human-AI Initiatives for Your Business
Two in danger after IBM's Watson supercomputer! Championed in 2011, artificial intelligence seemed ready to take on the world's biggest challenges. Indeed, after the victory of Watson's widely publicized quiz show, IBM teamed up with some of the leading US medical institutions to use Watson's algorithms to tackle the cancer crisis. It was expected that Watson's AI. The capabilities will analyze the vast amounts of cancer data that institutions collect, develop data-driven insights, and help care providers make more effective treatment decisions.
The initiative did not go according to plan. Oncologists turned to AI for answers, but Watson couldn't deliver for a variety of reasons, including gaps and glitches in data and the AI's inability to pick up on text cues in medical documents that are clear to doctors. Many of the Cancer Initiative projects eventually closed.
But in some projects oncologists and engineers took a different step. Instead of blaming the AI. In order not to produce results, they redesigned the respective roles of humans and algorithms. They realized that AI Genes mentioned in thousands of digitized academic papers can rapidly cross-reference a patient's genetic profile against mutations and identify treatments that may have been overlooked. Instead of asking Watson for a solution, he asked for an alternative to the solution. In reimagining the respective roles of oncologist and AI, both were able to play to their strengths: AI that reduced the time and effort required to identify a comprehensive list of potential treatments, while oncologists used their experience to choose among those treatments and deliver them to patients.
As businesses increasingly adopt AI, company leaders should heed Watson's lesson. In order to generate optimal results from their investment in AI, they must understand the different ways in which employees and algorithms can be combined and create the most effective human-AI. Combinations for the challenge at hand. Here are three principles for making this happen.
READ MORE: Artificial Intelligence helps decipher the sounds of the animal kingdom | Technology
First, the human-AI understand combinations available for your company
Our studies show that there are four templates leaders can follow as they combine employees and AI, as exemplified by Australian mining giant Rio Tinto.
AI As Publisher
AI Generates data-driven insights to expand the breadth and depth of employee thinking. Rio Tinto uses algorithms to analyze thousands of drill-core logs and create 3-D models of ore bodies. This allows exploration teams to better understand the structural configuration of composite ore deposits, which opens up new avenues in the exploration of mineral resources.
AI As Recommender
AI provides recommendations and employees decide which ones to act. Rio Tinto uses AI to make predictive maintenance recommendations that its experts use to create appliance repair programs.
AI As a Judge
AI makes decisions and employees implement them. Rio Tinto applies machine learning and mathematical programming to make real-time decisions to dispatch load-hall-dump (LHD) machines to its Argyle Diamond Mine in Australia; Those decisions are then made by human operators.
AI As Automator: AI Makes and executes decisions with employee oversight. Rio Tinto's Pilbara mines in Western Australia has minimal on-site workers, with most of its engineers overseeing its operations from as far away as Perth, 900 miles away. Fully automated, driverless trains carry 28,000 tons of iron ore from the mine to the port 200 miles away.
Second, the best Human-AI use a decision tree to determine combination for your application
As Rio Tinto demonstrates, the staff and AI. There is no single combination of. Naturally one is superior to the other. Instead, the officers were given healthy Human-AI Combination Choice: By asking yourself questions that explain the objectives, context, and their expected results. Together, these questions asked in sequence form a decision tree.
Objective
Do we want to implement a new business model, or improve the efficiency of an existing process? If the former, consider using AI. As publisher, that will help encourage creativity. If it's the latter, then other combinations (recommendator, decision maker and automator) are better choices.
Context
Do we have data that answers the questions an employee would ask when solving a problem, and if so, whether AI. What can be trained to answer those questions using the data? If the data is available and can be used to train AI, consider using AI. As a decider or an automaker. If the data is not available or cannot be used for AI training, use AI as a recommender.
Result
Can deploying AI deliver better results than deploying employees? (Typically, this is the case for large-scale, routine processes that require rapid execution.) If the answer to the question is "yes," consider using AI. as the Automator. If not, use AI only as a conclusive.
In addition to these questions, No matter what Human-AI as combination companies choose, they must also earn a social license to use AI by designing fair and transparent algorithms; Assuring stakeholders that AI higher cost; And proving that with data acquisition they can be trusted and accountable for their AI's decisions. In addition, as A.I. As its role evolves from illuminator to automator, companies will face ever higher barriers to gaining society's acceptance, with decision-making and formerly taking on a large portion of the implementation carried out by employees.
Domino's Pizza offers a nice, after-the-fact demonstration of how a decision tree can work. When customers in Australia complained that Domino's products "don't look good", the company called AI. It developed the DOM Pizza Checker, a scanner equipped with computer vision, that compared each pizza made by employees to a database of ideal pizza pictures to make sure it was appealing.
Domino's aimed to optimize the output of the existing pizza-making process, and an appropriately trained algorithm was able to do so by checking for the presence of the pizza (context). Pizza making is a fast-paced and iterative process that must ensure product consistency at each store; Hence, more time and cost can be saved by using AI-powered robots to execute the entire process(s). But it must be the employees who create (and fix) products in line with the company's mission of delivering quality handmade pizza: Automating the entire pizza-making process would jeopardize Domino's social license. This line of reasoning inspired Domino Australia to use AI as decisive.
Third, regularly use your Human-AI Review and refine it
Combination, employees and AI Ideal mix of Develops with technology and company-specific factors, such as objectives, capabilities and risk tolerance. Thus, the authorities must revisit the decision tree from time to time to find out if their human-a.k.a. The combination is still optimal.
The aircraft's engine-makers were Rolls-Royce Digital Twins and AI uses it. To make preventive maintenance recommendations to its customers. Over time its Engine Health Monitoring (EHM) system has become more sophisticated, measuring more performance parameters and feeding a richer data stream into its algorithms. As this trend continues, we expect the EHM system to evolve from a recommendation to decision system, telling airlines when and where to service specific engine parts.
Source: François Candelon, Fortune, Direct News 99
Post a Comment